diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-F16.md b/scores/Dolphin-Mistral-24B-Venice-Edition-F16.md new file mode 100644 index 0000000..45c52a4 --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-F16.md @@ -0,0 +1,1206 @@ +# Dolphin-Mistral-24B-Venice-Edition-F16.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 42 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 363 | +| 3 | UINT64 | 1 | GGUF.kv_count | 39 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.block_count | 40 | +| 17 | UINT32 | 1 | llama.context_length | 32768 | +| 18 | UINT32 | 1 | llama.embedding_length | 5120 | +| 19 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 20 | UINT32 | 1 | llama.attention.head_count | 32 | +| 21 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 22 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 23 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 24 | UINT32 | 1 | llama.attention.key_length | 128 | +| 25 | UINT32 | 1 | llama.attention.value_length | 128 | +| 26 | UINT32 | 1 | general.file_type | 1 | +| 27 | UINT32 | 1 | llama.vocab_size | 131072 | +| 28 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 29 | UINT32 | 1 | general.quantization_version | 2 | +| 30 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 31 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 32 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 33 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 34 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 35 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 36 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 37 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 38 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 39 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 40 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 41 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 42 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | + +## Tensors Overview ~24B Elements + +Total number of elements in all tensors: 23572403200 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-F16.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-F16gguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~24B Elements](#tensors-overview-24b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + - [Block 38 Tensor Group : ~556M Elements](#block-38-tensor-group--556m-elements) + - [Block 39 Tensor Group : ~556M Elements](#block-39-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784800 | 0x50000000 | +| 1 | token_embd.weight | 0x50784800 | 0x50000000 | +| 2 | blk.0.attn_norm.weight | 0xa0784800 | 0x5000 | +| 3 | blk.0.ffn_down.weight | 0xa0789800 | 0x14000000 | +| 4 | blk.0.ffn_gate.weight | 0xb4789800 | 0x14000000 | +| 5 | blk.0.ffn_up.weight | 0xc8789800 | 0x14000000 | +| 6 | blk.0.ffn_norm.weight | 0xdc789800 | 0x5000 | +| 7 | blk.0.attn_k.weight | 0xdc78e800 | 0xa00000 | +| 8 | blk.0.attn_output.weight | 0xdd18e800 | 0x2800000 | +| 9 | blk.0.attn_q.weight | 0xdf98e800 | 0x2800000 | +| 10 | blk.0.attn_v.weight | 0xe218e800 | 0xa00000 | +| 11 | blk.1.attn_norm.weight | 0xe2b8e800 | 0x5000 | +| 12 | blk.1.ffn_down.weight | 0xe2b93800 | 0x14000000 | +| 13 | blk.1.ffn_gate.weight | 0xf6b93800 | 0x14000000 | +| 14 | blk.1.ffn_up.weight | 0x10ab93800 | 0x14000000 | +| 15 | blk.1.ffn_norm.weight | 0x11eb93800 | 0x5000 | +| 16 | blk.1.attn_k.weight | 0x11eb98800 | 0xa00000 | +| 17 | blk.1.attn_output.weight | 0x11f598800 | 0x2800000 | +| 18 | blk.1.attn_q.weight | 0x121d98800 | 0x2800000 | +| 19 | blk.1.attn_v.weight | 0x124598800 | 0xa00000 | +| 20 | blk.10.attn_norm.weight | 0x124f98800 | 0x5000 | +| 21 | blk.10.ffn_down.weight | 0x124f9d800 | 0x14000000 | +| 22 | blk.10.ffn_gate.weight | 0x138f9d800 | 0x14000000 | +| 23 | blk.10.ffn_up.weight | 0x14cf9d800 | 0x14000000 | +| 24 | blk.10.ffn_norm.weight | 0x160f9d800 | 0x5000 | +| 25 | blk.10.attn_k.weight | 0x160fa2800 | 0xa00000 | +| 26 | blk.10.attn_output.weight | 0x1619a2800 | 0x2800000 | +| 27 | blk.10.attn_q.weight | 0x1641a2800 | 0x2800000 | +| 28 | blk.10.attn_v.weight | 0x1669a2800 | 0xa00000 | +| 29 | blk.11.attn_norm.weight | 0x1673a2800 | 0x5000 | +| 30 | blk.11.ffn_down.weight | 0x1673a7800 | 0x14000000 | +| 31 | blk.11.ffn_gate.weight | 0x17b3a7800 | 0x14000000 | +| 32 | blk.11.ffn_up.weight | 0x18f3a7800 | 0x14000000 | +| 33 | blk.11.ffn_norm.weight | 0x1a33a7800 | 0x5000 | +| 34 | blk.11.attn_k.weight | 0x1a33ac800 | 0xa00000 | +| 35 | blk.11.attn_output.weight | 0x1a3dac800 | 0x2800000 | +| 36 | blk.11.attn_q.weight | 0x1a65ac800 | 0x2800000 | +| 37 | blk.11.attn_v.weight | 0x1a8dac800 | 0xa00000 | +| 38 | blk.12.attn_norm.weight | 0x1a97ac800 | 0x5000 | +| 39 | blk.12.ffn_down.weight | 0x1a97b1800 | 0x14000000 | +| 40 | blk.12.ffn_gate.weight | 0x1bd7b1800 | 0x14000000 | +| 41 | blk.12.ffn_up.weight | 0x1d17b1800 | 0x14000000 | +| 42 | blk.12.ffn_norm.weight | 0x1e57b1800 | 0x5000 | +| 43 | blk.12.attn_k.weight | 0x1e57b6800 | 0xa00000 | +| 44 | blk.12.attn_output.weight | 0x1e61b6800 | 0x2800000 | +| 45 | blk.12.attn_q.weight | 0x1e89b6800 | 0x2800000 | +| 46 | blk.12.attn_v.weight | 0x1eb1b6800 | 0xa00000 | +| 47 | blk.13.attn_norm.weight | 0x1ebbb6800 | 0x5000 | +| 48 | blk.13.ffn_down.weight | 0x1ebbbb800 | 0x14000000 | +| 49 | blk.13.ffn_gate.weight | 0x1ffbbb800 | 0x14000000 | +| 50 | blk.13.ffn_up.weight | 0x213bbb800 | 0x14000000 | +| 51 | blk.13.ffn_norm.weight | 0x227bbb800 | 0x5000 | +| 52 | blk.13.attn_k.weight | 0x227bc0800 | 0xa00000 | +| 53 | blk.13.attn_output.weight | 0x2285c0800 | 0x2800000 | +| 54 | blk.13.attn_q.weight | 0x22adc0800 | 0x2800000 | +| 55 | blk.13.attn_v.weight | 0x22d5c0800 | 0xa00000 | +| 56 | blk.14.attn_norm.weight | 0x22dfc0800 | 0x5000 | +| 57 | blk.14.ffn_down.weight | 0x22dfc5800 | 0x14000000 | +| 58 | blk.14.ffn_gate.weight | 0x241fc5800 | 0x14000000 | +| 59 | blk.14.ffn_up.weight | 0x255fc5800 | 0x14000000 | +| 60 | blk.14.ffn_norm.weight | 0x269fc5800 | 0x5000 | +| 61 | blk.14.attn_k.weight | 0x269fca800 | 0xa00000 | +| 62 | blk.14.attn_output.weight | 0x26a9ca800 | 0x2800000 | +| 63 | blk.14.attn_q.weight | 0x26d1ca800 | 0x2800000 | +| 64 | blk.14.attn_v.weight | 0x26f9ca800 | 0xa00000 | +| 65 | blk.15.attn_norm.weight | 0x2703ca800 | 0x5000 | +| 66 | blk.15.ffn_down.weight | 0x2703cf800 | 0x14000000 | +| 67 | blk.15.ffn_gate.weight | 0x2843cf800 | 0x14000000 | +| 68 | blk.15.ffn_up.weight | 0x2983cf800 | 0x14000000 | +| 69 | blk.15.ffn_norm.weight | 0x2ac3cf800 | 0x5000 | +| 70 | blk.15.attn_k.weight | 0x2ac3d4800 | 0xa00000 | +| 71 | blk.15.attn_output.weight | 0x2acdd4800 | 0x2800000 | +| 72 | blk.15.attn_q.weight | 0x2af5d4800 | 0x2800000 | +| 73 | blk.15.attn_v.weight | 0x2b1dd4800 | 0xa00000 | +| 74 | blk.16.attn_norm.weight | 0x2b27d4800 | 0x5000 | +| 75 | blk.16.ffn_down.weight | 0x2b27d9800 | 0x14000000 | +| 76 | blk.16.ffn_gate.weight | 0x2c67d9800 | 0x14000000 | +| 77 | blk.16.ffn_up.weight | 0x2da7d9800 | 0x14000000 | +| 78 | blk.16.ffn_norm.weight | 0x2ee7d9800 | 0x5000 | +| 79 | blk.16.attn_k.weight | 0x2ee7de800 | 0xa00000 | +| 80 | blk.16.attn_output.weight | 0x2ef1de800 | 0x2800000 | +| 81 | blk.16.attn_q.weight | 0x2f19de800 | 0x2800000 | +| 82 | blk.16.attn_v.weight | 0x2f41de800 | 0xa00000 | +| 83 | blk.17.attn_norm.weight | 0x2f4bde800 | 0x5000 | +| 84 | blk.17.ffn_down.weight | 0x2f4be3800 | 0x14000000 | +| 85 | blk.17.ffn_gate.weight | 0x308be3800 | 0x14000000 | +| 86 | blk.17.ffn_up.weight | 0x31cbe3800 | 0x14000000 | +| 87 | blk.17.ffn_norm.weight | 0x330be3800 | 0x5000 | +| 88 | blk.17.attn_k.weight | 0x330be8800 | 0xa00000 | +| 89 | blk.17.attn_output.weight | 0x3315e8800 | 0x2800000 | +| 90 | blk.17.attn_q.weight | 0x333de8800 | 0x2800000 | +| 91 | blk.17.attn_v.weight | 0x3365e8800 | 0xa00000 | +| 92 | blk.18.attn_norm.weight | 0x336fe8800 | 0x5000 | +| 93 | blk.18.ffn_down.weight | 0x336fed800 | 0x14000000 | +| 94 | blk.18.ffn_gate.weight | 0x34afed800 | 0x14000000 | +| 95 | blk.18.ffn_up.weight | 0x35efed800 | 0x14000000 | +| 96 | blk.18.ffn_norm.weight | 0x372fed800 | 0x5000 | +| 97 | blk.18.attn_k.weight | 0x372ff2800 | 0xa00000 | +| 98 | blk.18.attn_output.weight | 0x3739f2800 | 0x2800000 | +| 99 | blk.18.attn_q.weight | 0x3761f2800 | 0x2800000 | +| 100 | blk.18.attn_v.weight | 0x3789f2800 | 0xa00000 | +| 101 | blk.19.attn_norm.weight | 0x3793f2800 | 0x5000 | +| 102 | blk.19.ffn_down.weight | 0x3793f7800 | 0x14000000 | +| 103 | blk.19.ffn_gate.weight | 0x38d3f7800 | 0x14000000 | +| 104 | blk.19.ffn_up.weight | 0x3a13f7800 | 0x14000000 | +| 105 | blk.19.ffn_norm.weight | 0x3b53f7800 | 0x5000 | +| 106 | blk.19.attn_k.weight | 0x3b53fc800 | 0xa00000 | +| 107 | blk.19.attn_output.weight | 0x3b5dfc800 | 0x2800000 | +| 108 | blk.19.attn_q.weight | 0x3b85fc800 | 0x2800000 | +| 109 | blk.19.attn_v.weight | 0x3badfc800 | 0xa00000 | +| 110 | blk.2.attn_norm.weight | 0x3bb7fc800 | 0x5000 | +| 111 | blk.2.ffn_down.weight | 0x3bb801800 | 0x14000000 | +| 112 | blk.2.ffn_gate.weight | 0x3cf801800 | 0x14000000 | +| 113 | blk.2.ffn_up.weight | 0x3e3801800 | 0x14000000 | +| 114 | blk.2.ffn_norm.weight | 0x3f7801800 | 0x5000 | +| 115 | blk.2.attn_k.weight | 0x3f7806800 | 0xa00000 | +| 116 | blk.2.attn_output.weight | 0x3f8206800 | 0x2800000 | +| 117 | blk.2.attn_q.weight | 0x3faa06800 | 0x2800000 | +| 118 | blk.2.attn_v.weight | 0x3fd206800 | 0xa00000 | +| 119 | blk.20.attn_norm.weight | 0x3fdc06800 | 0x5000 | +| 120 | blk.20.ffn_down.weight | 0x3fdc0b800 | 0x14000000 | +| 121 | blk.20.ffn_gate.weight | 0x411c0b800 | 0x14000000 | +| 122 | blk.20.ffn_up.weight | 0x425c0b800 | 0x14000000 | +| 123 | blk.20.ffn_norm.weight | 0x439c0b800 | 0x5000 | +| 124 | blk.20.attn_k.weight | 0x439c10800 | 0xa00000 | +| 125 | blk.20.attn_output.weight | 0x43a610800 | 0x2800000 | +| 126 | blk.20.attn_q.weight | 0x43ce10800 | 0x2800000 | +| 127 | blk.20.attn_v.weight | 0x43f610800 | 0xa00000 | +| 128 | blk.21.attn_norm.weight | 0x440010800 | 0x5000 | +| 129 | blk.21.ffn_down.weight | 0x440015800 | 0x14000000 | +| 130 | blk.21.ffn_gate.weight | 0x454015800 | 0x14000000 | +| 131 | blk.21.ffn_up.weight | 0x468015800 | 0x14000000 | +| 132 | blk.21.ffn_norm.weight | 0x47c015800 | 0x5000 | +| 133 | blk.21.attn_k.weight | 0x47c01a800 | 0xa00000 | +| 134 | blk.21.attn_output.weight | 0x47ca1a800 | 0x2800000 | +| 135 | blk.21.attn_q.weight | 0x47f21a800 | 0x2800000 | +| 136 | blk.21.attn_v.weight | 0x481a1a800 | 0xa00000 | +| 137 | blk.22.attn_norm.weight | 0x48241a800 | 0x5000 | +| 138 | blk.22.ffn_down.weight | 0x48241f800 | 0x14000000 | +| 139 | blk.22.ffn_gate.weight | 0x49641f800 | 0x14000000 | +| 140 | blk.22.ffn_up.weight | 0x4aa41f800 | 0x14000000 | +| 141 | blk.22.ffn_norm.weight | 0x4be41f800 | 0x5000 | +| 142 | blk.22.attn_k.weight | 0x4be424800 | 0xa00000 | +| 143 | blk.22.attn_output.weight | 0x4bee24800 | 0x2800000 | +| 144 | blk.22.attn_q.weight | 0x4c1624800 | 0x2800000 | +| 145 | blk.22.attn_v.weight | 0x4c3e24800 | 0xa00000 | +| 146 | blk.23.attn_norm.weight | 0x4c4824800 | 0x5000 | +| 147 | blk.23.ffn_down.weight | 0x4c4829800 | 0x14000000 | +| 148 | blk.23.ffn_gate.weight | 0x4d8829800 | 0x14000000 | +| 149 | blk.23.ffn_up.weight | 0x4ec829800 | 0x14000000 | +| 150 | blk.23.ffn_norm.weight | 0x500829800 | 0x5000 | +| 151 | blk.23.attn_k.weight | 0x50082e800 | 0xa00000 | +| 152 | blk.23.attn_output.weight | 0x50122e800 | 0x2800000 | +| 153 | blk.23.attn_q.weight | 0x503a2e800 | 0x2800000 | +| 154 | blk.23.attn_v.weight | 0x50622e800 | 0xa00000 | +| 155 | blk.24.attn_norm.weight | 0x506c2e800 | 0x5000 | +| 156 | blk.24.ffn_down.weight | 0x506c33800 | 0x14000000 | +| 157 | blk.24.ffn_gate.weight | 0x51ac33800 | 0x14000000 | +| 158 | blk.24.ffn_up.weight | 0x52ec33800 | 0x14000000 | +| 159 | blk.24.ffn_norm.weight | 0x542c33800 | 0x5000 | +| 160 | blk.24.attn_k.weight | 0x542c38800 | 0xa00000 | +| 161 | blk.24.attn_output.weight | 0x543638800 | 0x2800000 | +| 162 | blk.24.attn_q.weight | 0x545e38800 | 0x2800000 | +| 163 | blk.24.attn_v.weight | 0x548638800 | 0xa00000 | +| 164 | blk.25.attn_norm.weight | 0x549038800 | 0x5000 | +| 165 | blk.25.ffn_down.weight | 0x54903d800 | 0x14000000 | +| 166 | blk.25.ffn_gate.weight | 0x55d03d800 | 0x14000000 | +| 167 | blk.25.ffn_up.weight | 0x57103d800 | 0x14000000 | +| 168 | blk.25.ffn_norm.weight | 0x58503d800 | 0x5000 | +| 169 | blk.25.attn_k.weight | 0x585042800 | 0xa00000 | +| 170 | blk.25.attn_output.weight | 0x585a42800 | 0x2800000 | +| 171 | blk.25.attn_q.weight | 0x588242800 | 0x2800000 | +| 172 | blk.25.attn_v.weight | 0x58aa42800 | 0xa00000 | +| 173 | blk.26.attn_norm.weight | 0x58b442800 | 0x5000 | +| 174 | blk.26.ffn_down.weight | 0x58b447800 | 0x14000000 | +| 175 | blk.26.ffn_gate.weight | 0x59f447800 | 0x14000000 | +| 176 | blk.26.ffn_up.weight | 0x5b3447800 | 0x14000000 | +| 177 | blk.26.ffn_norm.weight | 0x5c7447800 | 0x5000 | +| 178 | blk.26.attn_k.weight | 0x5c744c800 | 0xa00000 | +| 179 | blk.26.attn_output.weight | 0x5c7e4c800 | 0x2800000 | +| 180 | blk.26.attn_q.weight | 0x5ca64c800 | 0x2800000 | +| 181 | blk.26.attn_v.weight | 0x5cce4c800 | 0xa00000 | +| 182 | blk.27.attn_norm.weight | 0x5cd84c800 | 0x5000 | +| 183 | blk.27.ffn_down.weight | 0x5cd851800 | 0x14000000 | +| 184 | blk.27.ffn_gate.weight | 0x5e1851800 | 0x14000000 | +| 185 | blk.27.ffn_up.weight | 0x5f5851800 | 0x14000000 | +| 186 | blk.27.ffn_norm.weight | 0x609851800 | 0x5000 | +| 187 | blk.27.attn_k.weight | 0x609856800 | 0xa00000 | +| 188 | blk.27.attn_output.weight | 0x60a256800 | 0x2800000 | +| 189 | blk.27.attn_q.weight | 0x60ca56800 | 0x2800000 | +| 190 | blk.27.attn_v.weight | 0x60f256800 | 0xa00000 | +| 191 | blk.28.attn_norm.weight | 0x60fc56800 | 0x5000 | +| 192 | blk.28.ffn_down.weight | 0x60fc5b800 | 0x14000000 | +| 193 | blk.28.ffn_gate.weight | 0x623c5b800 | 0x14000000 | +| 194 | blk.28.ffn_up.weight | 0x637c5b800 | 0x14000000 | +| 195 | blk.28.ffn_norm.weight | 0x64bc5b800 | 0x5000 | +| 196 | blk.28.attn_k.weight | 0x64bc60800 | 0xa00000 | +| 197 | blk.28.attn_output.weight | 0x64c660800 | 0x2800000 | +| 198 | blk.28.attn_q.weight | 0x64ee60800 | 0x2800000 | +| 199 | blk.28.attn_v.weight | 0x651660800 | 0xa00000 | +| 200 | blk.29.attn_norm.weight | 0x652060800 | 0x5000 | +| 201 | blk.29.ffn_down.weight | 0x652065800 | 0x14000000 | +| 202 | blk.29.ffn_gate.weight | 0x666065800 | 0x14000000 | +| 203 | blk.29.ffn_up.weight | 0x67a065800 | 0x14000000 | +| 204 | blk.29.ffn_norm.weight | 0x68e065800 | 0x5000 | +| 205 | blk.29.attn_k.weight | 0x68e06a800 | 0xa00000 | +| 206 | blk.29.attn_output.weight | 0x68ea6a800 | 0x2800000 | +| 207 | blk.29.attn_q.weight | 0x69126a800 | 0x2800000 | +| 208 | blk.29.attn_v.weight | 0x693a6a800 | 0xa00000 | +| 209 | blk.3.attn_norm.weight | 0x69446a800 | 0x5000 | +| 210 | blk.3.ffn_down.weight | 0x69446f800 | 0x14000000 | +| 211 | blk.3.ffn_gate.weight | 0x6a846f800 | 0x14000000 | +| 212 | blk.3.ffn_up.weight | 0x6bc46f800 | 0x14000000 | +| 213 | blk.3.ffn_norm.weight | 0x6d046f800 | 0x5000 | +| 214 | blk.3.attn_k.weight | 0x6d0474800 | 0xa00000 | +| 215 | blk.3.attn_output.weight | 0x6d0e74800 | 0x2800000 | +| 216 | blk.3.attn_q.weight | 0x6d3674800 | 0x2800000 | +| 217 | blk.3.attn_v.weight | 0x6d5e74800 | 0xa00000 | +| 218 | blk.30.attn_norm.weight | 0x6d6874800 | 0x5000 | +| 219 | blk.30.ffn_down.weight | 0x6d6879800 | 0x14000000 | +| 220 | blk.30.ffn_gate.weight | 0x6ea879800 | 0x14000000 | +| 221 | blk.30.ffn_up.weight | 0x6fe879800 | 0x14000000 | +| 222 | blk.30.ffn_norm.weight | 0x712879800 | 0x5000 | +| 223 | blk.30.attn_k.weight | 0x71287e800 | 0xa00000 | +| 224 | blk.30.attn_output.weight | 0x71327e800 | 0x2800000 | +| 225 | blk.30.attn_q.weight | 0x715a7e800 | 0x2800000 | +| 226 | blk.30.attn_v.weight | 0x71827e800 | 0xa00000 | +| 227 | blk.31.attn_norm.weight | 0x718c7e800 | 0x5000 | +| 228 | blk.31.ffn_down.weight | 0x718c83800 | 0x14000000 | +| 229 | blk.31.ffn_gate.weight | 0x72cc83800 | 0x14000000 | +| 230 | blk.31.ffn_up.weight | 0x740c83800 | 0x14000000 | +| 231 | blk.31.ffn_norm.weight | 0x754c83800 | 0x5000 | +| 232 | blk.31.attn_k.weight | 0x754c88800 | 0xa00000 | +| 233 | blk.31.attn_output.weight | 0x755688800 | 0x2800000 | +| 234 | blk.31.attn_q.weight | 0x757e88800 | 0x2800000 | +| 235 | blk.31.attn_v.weight | 0x75a688800 | 0xa00000 | +| 236 | blk.32.attn_norm.weight | 0x75b088800 | 0x5000 | +| 237 | blk.32.ffn_down.weight | 0x75b08d800 | 0x14000000 | +| 238 | blk.32.ffn_gate.weight | 0x76f08d800 | 0x14000000 | +| 239 | blk.32.ffn_up.weight | 0x78308d800 | 0x14000000 | +| 240 | blk.32.ffn_norm.weight | 0x79708d800 | 0x5000 | +| 241 | blk.32.attn_k.weight | 0x797092800 | 0xa00000 | +| 242 | blk.32.attn_output.weight | 0x797a92800 | 0x2800000 | +| 243 | blk.32.attn_q.weight | 0x79a292800 | 0x2800000 | +| 244 | blk.32.attn_v.weight | 0x79ca92800 | 0xa00000 | +| 245 | blk.33.attn_norm.weight | 0x79d492800 | 0x5000 | +| 246 | blk.33.ffn_down.weight | 0x79d497800 | 0x14000000 | +| 247 | blk.33.ffn_gate.weight | 0x7b1497800 | 0x14000000 | +| 248 | blk.33.ffn_up.weight | 0x7c5497800 | 0x14000000 | +| 249 | blk.33.ffn_norm.weight | 0x7d9497800 | 0x5000 | +| 250 | blk.33.attn_k.weight | 0x7d949c800 | 0xa00000 | +| 251 | blk.33.attn_output.weight | 0x7d9e9c800 | 0x2800000 | +| 252 | blk.33.attn_q.weight | 0x7dc69c800 | 0x2800000 | +| 253 | blk.33.attn_v.weight | 0x7dee9c800 | 0xa00000 | +| 254 | blk.34.attn_norm.weight | 0x7df89c800 | 0x5000 | +| 255 | blk.34.ffn_down.weight | 0x7df8a1800 | 0x14000000 | +| 256 | blk.34.ffn_gate.weight | 0x7f38a1800 | 0x14000000 | +| 257 | blk.34.ffn_up.weight | 0x8078a1800 | 0x14000000 | +| 258 | blk.34.ffn_norm.weight | 0x81b8a1800 | 0x5000 | +| 259 | blk.34.attn_k.weight | 0x81b8a6800 | 0xa00000 | +| 260 | blk.34.attn_output.weight | 0x81c2a6800 | 0x2800000 | +| 261 | blk.34.attn_q.weight | 0x81eaa6800 | 0x2800000 | +| 262 | blk.34.attn_v.weight | 0x8212a6800 | 0xa00000 | +| 263 | blk.35.attn_norm.weight | 0x821ca6800 | 0x5000 | +| 264 | blk.35.ffn_down.weight | 0x821cab800 | 0x14000000 | +| 265 | blk.35.ffn_gate.weight | 0x835cab800 | 0x14000000 | +| 266 | blk.35.ffn_up.weight | 0x849cab800 | 0x14000000 | +| 267 | blk.35.ffn_norm.weight | 0x85dcab800 | 0x5000 | +| 268 | blk.35.attn_k.weight | 0x85dcb0800 | 0xa00000 | +| 269 | blk.35.attn_output.weight | 0x85e6b0800 | 0x2800000 | +| 270 | blk.35.attn_q.weight | 0x860eb0800 | 0x2800000 | +| 271 | blk.35.attn_v.weight | 0x8636b0800 | 0xa00000 | +| 272 | blk.36.attn_norm.weight | 0x8640b0800 | 0x5000 | +| 273 | blk.36.ffn_down.weight | 0x8640b5800 | 0x14000000 | +| 274 | blk.36.ffn_gate.weight | 0x8780b5800 | 0x14000000 | +| 275 | blk.36.ffn_up.weight | 0x88c0b5800 | 0x14000000 | +| 276 | blk.36.ffn_norm.weight | 0x8a00b5800 | 0x5000 | +| 277 | blk.36.attn_k.weight | 0x8a00ba800 | 0xa00000 | +| 278 | blk.36.attn_output.weight | 0x8a0aba800 | 0x2800000 | +| 279 | blk.36.attn_q.weight | 0x8a32ba800 | 0x2800000 | +| 280 | blk.36.attn_v.weight | 0x8a5aba800 | 0xa00000 | +| 281 | blk.37.attn_norm.weight | 0x8a64ba800 | 0x5000 | +| 282 | blk.37.ffn_down.weight | 0x8a64bf800 | 0x14000000 | +| 283 | blk.37.ffn_gate.weight | 0x8ba4bf800 | 0x14000000 | +| 284 | blk.37.ffn_up.weight | 0x8ce4bf800 | 0x14000000 | +| 285 | blk.37.ffn_norm.weight | 0x8e24bf800 | 0x5000 | +| 286 | blk.37.attn_k.weight | 0x8e24c4800 | 0xa00000 | +| 287 | blk.37.attn_output.weight | 0x8e2ec4800 | 0x2800000 | +| 288 | blk.37.attn_q.weight | 0x8e56c4800 | 0x2800000 | +| 289 | blk.37.attn_v.weight | 0x8e7ec4800 | 0xa00000 | +| 290 | blk.38.attn_norm.weight | 0x8e88c4800 | 0x5000 | +| 291 | blk.38.ffn_down.weight | 0x8e88c9800 | 0x14000000 | +| 292 | blk.38.ffn_gate.weight | 0x8fc8c9800 | 0x14000000 | +| 293 | blk.38.ffn_up.weight | 0x9108c9800 | 0x14000000 | +| 294 | blk.38.ffn_norm.weight | 0x9248c9800 | 0x5000 | +| 295 | blk.38.attn_k.weight | 0x9248ce800 | 0xa00000 | +| 296 | blk.38.attn_output.weight | 0x9252ce800 | 0x2800000 | +| 297 | blk.38.attn_q.weight | 0x927ace800 | 0x2800000 | +| 298 | blk.38.attn_v.weight | 0x92a2ce800 | 0xa00000 | +| 299 | blk.39.attn_norm.weight | 0x92acce800 | 0x5000 | +| 300 | blk.39.ffn_down.weight | 0x92acd3800 | 0x14000000 | +| 301 | blk.39.ffn_gate.weight | 0x93ecd3800 | 0x14000000 | +| 302 | blk.39.ffn_up.weight | 0x952cd3800 | 0x14000000 | +| 303 | blk.39.ffn_norm.weight | 0x966cd3800 | 0x5000 | +| 304 | blk.39.attn_k.weight | 0x966cd8800 | 0xa00000 | +| 305 | blk.39.attn_output.weight | 0x9676d8800 | 0x2800000 | +| 306 | blk.39.attn_q.weight | 0x969ed8800 | 0x2800000 | +| 307 | blk.39.attn_v.weight | 0x96c6d8800 | 0xa00000 | +| 308 | blk.4.attn_norm.weight | 0x96d0d8800 | 0x5000 | +| 309 | blk.4.ffn_down.weight | 0x96d0dd800 | 0x14000000 | +| 310 | blk.4.ffn_gate.weight | 0x9810dd800 | 0x14000000 | +| 311 | blk.4.ffn_up.weight | 0x9950dd800 | 0x14000000 | +| 312 | blk.4.ffn_norm.weight | 0x9a90dd800 | 0x5000 | +| 313 | blk.4.attn_k.weight | 0x9a90e2800 | 0xa00000 | +| 314 | blk.4.attn_output.weight | 0x9a9ae2800 | 0x2800000 | +| 315 | blk.4.attn_q.weight | 0x9ac2e2800 | 0x2800000 | +| 316 | blk.4.attn_v.weight | 0x9aeae2800 | 0xa00000 | +| 317 | blk.5.attn_norm.weight | 0x9af4e2800 | 0x5000 | +| 318 | blk.5.ffn_down.weight | 0x9af4e7800 | 0x14000000 | +| 319 | blk.5.ffn_gate.weight | 0x9c34e7800 | 0x14000000 | +| 320 | blk.5.ffn_up.weight | 0x9d74e7800 | 0x14000000 | +| 321 | blk.5.ffn_norm.weight | 0x9eb4e7800 | 0x5000 | +| 322 | blk.5.attn_k.weight | 0x9eb4ec800 | 0xa00000 | +| 323 | blk.5.attn_output.weight | 0x9ebeec800 | 0x2800000 | +| 324 | blk.5.attn_q.weight | 0x9ee6ec800 | 0x2800000 | +| 325 | blk.5.attn_v.weight | 0x9f0eec800 | 0xa00000 | +| 326 | blk.6.attn_norm.weight | 0x9f18ec800 | 0x5000 | +| 327 | blk.6.ffn_down.weight | 0x9f18f1800 | 0x14000000 | +| 328 | blk.6.ffn_gate.weight | 0xa058f1800 | 0x14000000 | +| 329 | blk.6.ffn_up.weight | 0xa198f1800 | 0x14000000 | +| 330 | blk.6.ffn_norm.weight | 0xa2d8f1800 | 0x5000 | +| 331 | blk.6.attn_k.weight | 0xa2d8f6800 | 0xa00000 | +| 332 | blk.6.attn_output.weight | 0xa2e2f6800 | 0x2800000 | +| 333 | blk.6.attn_q.weight | 0xa30af6800 | 0x2800000 | +| 334 | blk.6.attn_v.weight | 0xa332f6800 | 0xa00000 | +| 335 | blk.7.attn_norm.weight | 0xa33cf6800 | 0x5000 | +| 336 | blk.7.ffn_down.weight | 0xa33cfb800 | 0x14000000 | +| 337 | blk.7.ffn_gate.weight | 0xa47cfb800 | 0x14000000 | +| 338 | blk.7.ffn_up.weight | 0xa5bcfb800 | 0x14000000 | +| 339 | blk.7.ffn_norm.weight | 0xa6fcfb800 | 0x5000 | +| 340 | blk.7.attn_k.weight | 0xa6fd00800 | 0xa00000 | +| 341 | blk.7.attn_output.weight | 0xa70700800 | 0x2800000 | +| 342 | blk.7.attn_q.weight | 0xa72f00800 | 0x2800000 | +| 343 | blk.7.attn_v.weight | 0xa75700800 | 0xa00000 | +| 344 | blk.8.attn_norm.weight | 0xa76100800 | 0x5000 | +| 345 | blk.8.ffn_down.weight | 0xa76105800 | 0x14000000 | +| 346 | blk.8.ffn_gate.weight | 0xa8a105800 | 0x14000000 | +| 347 | blk.8.ffn_up.weight | 0xa9e105800 | 0x14000000 | +| 348 | blk.8.ffn_norm.weight | 0xab2105800 | 0x5000 | +| 349 | blk.8.attn_k.weight | 0xab210a800 | 0xa00000 | +| 350 | blk.8.attn_output.weight | 0xab2b0a800 | 0x2800000 | +| 351 | blk.8.attn_q.weight | 0xab530a800 | 0x2800000 | +| 352 | blk.8.attn_v.weight | 0xab7b0a800 | 0xa00000 | +| 353 | blk.9.attn_norm.weight | 0xab850a800 | 0x5000 | +| 354 | blk.9.ffn_down.weight | 0xab850f800 | 0x14000000 | +| 355 | blk.9.ffn_gate.weight | 0xacc50f800 | 0x14000000 | +| 356 | blk.9.ffn_up.weight | 0xae050f800 | 0x14000000 | +| 357 | blk.9.ffn_norm.weight | 0xaf450f800 | 0x5000 | +| 358 | blk.9.attn_k.weight | 0xaf4514800 | 0xa00000 | +| 359 | blk.9.attn_output.weight | 0xaf4f14800 | 0x2800000 | +| 360 | blk.9.attn_q.weight | 0xaf7714800 | 0x2800000 | +| 361 | blk.9.attn_v.weight | 0xaf9f14800 | 0xa00000 | +| 362 | output_norm.weight | 0xafa914800 | 0x5000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | F16 | +| 1 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | F16 | +| 362 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.69% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 2 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 3 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 4 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 5 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 6 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 7 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 8 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 9 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 10 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 11 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 12 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 13 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 14 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 15 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 16 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 17 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 18 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 19 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 20 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 21 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 22 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 23 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 24 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 25 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 26 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 27 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 28 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 29 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 30 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 31 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 32 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 33 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 34 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 35 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 36 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 37 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 38 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 39 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 40 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 41 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 42 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 43 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 44 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 45 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 46 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 47 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 48 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 49 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 50 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 51 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 52 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 53 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 54 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 55 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 56 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 57 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 58 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 59 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 60 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 61 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 62 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 63 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 64 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 65 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 66 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 67 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 68 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 69 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 70 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 71 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 72 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 73 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 74 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 75 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 76 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 77 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 78 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 79 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 80 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 81 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 82 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 83 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 84 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 85 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 86 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 87 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 88 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 89 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 90 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 91 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 92 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 93 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 94 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 95 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 96 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 97 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 98 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 99 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 100 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 101 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 102 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 103 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 104 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 105 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 106 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 107 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 108 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 109 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 110 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 111 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 112 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 113 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 114 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 115 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 116 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 117 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 118 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 119 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 120 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 121 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 122 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 123 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 124 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 125 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 126 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 127 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 128 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 129 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 130 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 131 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 132 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 133 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 134 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 135 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 136 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 137 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 138 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 139 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 140 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 141 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 142 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 143 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 144 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 145 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 146 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 147 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 148 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 149 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 150 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 151 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 152 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 153 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 154 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 155 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 156 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 157 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 158 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 159 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 160 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 161 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 162 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 163 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 164 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 165 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 166 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 167 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 168 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 169 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 170 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 171 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 172 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 173 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 174 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 175 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 176 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 177 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 178 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 179 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 180 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 181 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 182 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 183 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 184 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 185 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 186 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 187 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 188 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 189 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 190 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 191 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 192 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 193 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 194 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 195 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 196 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 197 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 198 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 199 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 200 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 201 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 202 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 203 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 204 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 205 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 206 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 207 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 208 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 209 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 210 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 211 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 212 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 213 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 214 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 215 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 216 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 217 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 218 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 219 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 220 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 221 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 222 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 223 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 224 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 225 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 226 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 227 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 228 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 229 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 230 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 231 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 232 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 233 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 234 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 235 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 236 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 237 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 238 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 239 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 240 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 241 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 242 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 243 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 244 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 245 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 246 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 247 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 248 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 249 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 250 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 251 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 252 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 253 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 254 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 255 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 256 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 257 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 258 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 259 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 260 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 261 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 262 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 263 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 264 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 265 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 266 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 267 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 268 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 269 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 270 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 271 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 272 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 273 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 274 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 275 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 276 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 277 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 278 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 279 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 280 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 281 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 282 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 283 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 284 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 285 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 286 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 287 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 288 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 289 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 38 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 290 | blk.38.attn_norm.weight | Block 38 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 291 | blk.38.ffn_down.weight | Block 38 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 292 | blk.38.ffn_gate.weight | Block 38 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 293 | blk.38.ffn_up.weight | Block 38 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 294 | blk.38.ffn_norm.weight | Block 38 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 295 | blk.38.attn_k.weight | Block 38 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 296 | blk.38.attn_output.weight | Block 38 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 297 | blk.38.attn_q.weight | Block 38 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 298 | blk.38.attn_v.weight | Block 38 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.38: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 39 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 299 | blk.39.attn_norm.weight | Block 39 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 300 | blk.39.ffn_down.weight | Block 39 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 301 | blk.39.ffn_gate.weight | Block 39 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 302 | blk.39.ffn_up.weight | Block 39 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 303 | blk.39.ffn_norm.weight | Block 39 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 304 | blk.39.attn_k.weight | Block 39 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 305 | blk.39.attn_output.weight | Block 39 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 306 | blk.39.attn_q.weight | Block 39 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 307 | blk.39.attn_v.weight | Block 39 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.39: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 308 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 309 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 310 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 311 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 312 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 313 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 314 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 315 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 316 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 317 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 318 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 319 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 320 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 321 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 322 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 323 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 324 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 325 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 326 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 327 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 328 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 329 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 330 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 331 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 332 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 333 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 334 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 335 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 336 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 337 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 338 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 339 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 340 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 341 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 342 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 343 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 344 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 345 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 346 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 347 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 348 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 349 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 350 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 351 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 352 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.36% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 353 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 354 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | F16 | +| 355 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 356 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | F16 | +| 357 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 358 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 359 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | F16 | +| 360 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | F16 | +| 361 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.36% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-IQ3_M.md b/scores/Dolphin-Mistral-24B-Venice-Edition-IQ3_M.md new file mode 100644 index 0000000..dfafe57 --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-IQ3_M.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-IQ3_M.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 27 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-IQ3\_M.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-iq3_mgguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0x11300000 | +| 1 | output_norm.weight | 0x11a84500 | 0x5000 | +| 2 | token_embd.weight | 0x11a89500 | 0x11300000 | +| 3 | blk.0.attn_k.weight | 0x22d89500 | 0x1ea000 | +| 4 | blk.0.attn_norm.weight | 0x22f73500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x22f78500 | 0xb40000 | +| 6 | blk.0.attn_q.weight | 0x23ab8500 | 0x7a8000 | +| 7 | blk.0.attn_v.weight | 0x24260500 | 0x226000 | +| 8 | blk.0.ffn_down.weight | 0x24486500 | 0x5a00000 | +| 9 | blk.0.ffn_gate.weight | 0x29e86500 | 0x3d40000 | +| 10 | blk.0.ffn_norm.weight | 0x2dbc6500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x2dbcb500 | 0x3d40000 | +| 12 | blk.1.attn_k.weight | 0x3190b500 | 0x1ea000 | +| 13 | blk.1.attn_norm.weight | 0x31af5500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x31afa500 | 0xb40000 | +| 15 | blk.1.attn_q.weight | 0x3263a500 | 0x7a8000 | +| 16 | blk.1.attn_v.weight | 0x32de2500 | 0x226000 | +| 17 | blk.1.ffn_down.weight | 0x33008500 | 0x5a00000 | +| 18 | blk.1.ffn_gate.weight | 0x38a08500 | 0x3d40000 | +| 19 | blk.1.ffn_norm.weight | 0x3c748500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x3c74d500 | 0x3d40000 | +| 21 | blk.2.attn_k.weight | 0x4048d500 | 0x1ea000 | +| 22 | blk.2.attn_norm.weight | 0x40677500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x4067c500 | 0xb40000 | +| 24 | blk.2.attn_q.weight | 0x411bc500 | 0x7a8000 | +| 25 | blk.2.attn_v.weight | 0x41964500 | 0x226000 | +| 26 | blk.2.ffn_down.weight | 0x41b8a500 | 0x5a00000 | +| 27 | blk.2.ffn_gate.weight | 0x4758a500 | 0x3d40000 | +| 28 | blk.2.ffn_norm.weight | 0x4b2ca500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x4b2cf500 | 0x3d40000 | +| 30 | blk.3.attn_k.weight | 0x4f00f500 | 0x1ea000 | +| 31 | blk.3.attn_norm.weight | 0x4f1f9500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x4f1fe500 | 0xb40000 | +| 33 | blk.3.attn_q.weight | 0x4fd3e500 | 0x7a8000 | +| 34 | blk.3.attn_v.weight | 0x504e6500 | 0x226000 | +| 35 | blk.3.ffn_down.weight | 0x5070c500 | 0x5a00000 | +| 36 | blk.3.ffn_gate.weight | 0x5610c500 | 0x3d40000 | +| 37 | blk.3.ffn_norm.weight | 0x59e4c500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0x59e51500 | 0x3d40000 | +| 39 | blk.4.attn_k.weight | 0x5db91500 | 0x1ea000 | +| 40 | blk.4.attn_norm.weight | 0x5dd7b500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0x5dd80500 | 0xb40000 | +| 42 | blk.4.attn_q.weight | 0x5e8c0500 | 0x7a8000 | +| 43 | blk.4.attn_v.weight | 0x5f068500 | 0x226000 | +| 44 | blk.4.ffn_down.weight | 0x5f28e500 | 0x5a00000 | +| 45 | blk.4.ffn_gate.weight | 0x64c8e500 | 0x3d40000 | +| 46 | blk.4.ffn_norm.weight | 0x689ce500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0x689d3500 | 0x3d40000 | +| 48 | blk.5.attn_k.weight | 0x6c713500 | 0x1ea000 | +| 49 | blk.5.attn_norm.weight | 0x6c8fd500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0x6c902500 | 0xb40000 | +| 51 | blk.5.attn_q.weight | 0x6d442500 | 0x7a8000 | +| 52 | blk.5.attn_v.weight | 0x6dbea500 | 0x226000 | +| 53 | blk.5.ffn_down.weight | 0x6de10500 | 0x44c0000 | +| 54 | blk.5.ffn_gate.weight | 0x722d0500 | 0x3d40000 | +| 55 | blk.5.ffn_norm.weight | 0x76010500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0x76015500 | 0x3d40000 | +| 57 | blk.6.attn_k.weight | 0x79d55500 | 0x1ea000 | +| 58 | blk.6.attn_norm.weight | 0x79f3f500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0x79f44500 | 0xb40000 | +| 60 | blk.6.attn_q.weight | 0x7aa84500 | 0x7a8000 | +| 61 | blk.6.attn_v.weight | 0x7b22c500 | 0x226000 | +| 62 | blk.6.ffn_down.weight | 0x7b452500 | 0x44c0000 | +| 63 | blk.6.ffn_gate.weight | 0x7f912500 | 0x3d40000 | +| 64 | blk.6.ffn_norm.weight | 0x83652500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0x83657500 | 0x3d40000 | +| 66 | blk.7.attn_k.weight | 0x87397500 | 0x1ea000 | +| 67 | blk.7.attn_norm.weight | 0x87581500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0x87586500 | 0xb40000 | +| 69 | blk.7.attn_q.weight | 0x880c6500 | 0x7a8000 | +| 70 | blk.7.attn_v.weight | 0x8886e500 | 0x226000 | +| 71 | blk.7.ffn_down.weight | 0x88a94500 | 0x44c0000 | +| 72 | blk.7.ffn_gate.weight | 0x8cf54500 | 0x3d40000 | +| 73 | blk.7.ffn_norm.weight | 0x90c94500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0x90c99500 | 0x3d40000 | +| 75 | blk.8.attn_k.weight | 0x949d9500 | 0x1ea000 | +| 76 | blk.8.attn_norm.weight | 0x94bc3500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0x94bc8500 | 0xb40000 | +| 78 | blk.8.attn_q.weight | 0x95708500 | 0x7a8000 | +| 79 | blk.8.attn_v.weight | 0x95eb0500 | 0x226000 | +| 80 | blk.8.ffn_down.weight | 0x960d6500 | 0x44c0000 | +| 81 | blk.8.ffn_gate.weight | 0x9a596500 | 0x3d40000 | +| 82 | blk.8.ffn_norm.weight | 0x9e2d6500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0x9e2db500 | 0x3d40000 | +| 84 | blk.9.attn_k.weight | 0xa201b500 | 0x1ea000 | +| 85 | blk.9.attn_norm.weight | 0xa2205500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0xa220a500 | 0xb40000 | +| 87 | blk.9.attn_q.weight | 0xa2d4a500 | 0x7a8000 | +| 88 | blk.9.attn_v.weight | 0xa34f2500 | 0x226000 | +| 89 | blk.9.ffn_down.weight | 0xa3718500 | 0x44c0000 | +| 90 | blk.9.ffn_gate.weight | 0xa7bd8500 | 0x3d40000 | +| 91 | blk.9.ffn_norm.weight | 0xab918500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0xab91d500 | 0x3d40000 | +| 93 | blk.10.attn_k.weight | 0xaf65d500 | 0x1ea000 | +| 94 | blk.10.attn_norm.weight | 0xaf847500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0xaf84c500 | 0xb40000 | +| 96 | blk.10.attn_q.weight | 0xb038c500 | 0x7a8000 | +| 97 | blk.10.attn_v.weight | 0xb0b34500 | 0x226000 | +| 98 | blk.10.ffn_down.weight | 0xb0d5a500 | 0x44c0000 | +| 99 | blk.10.ffn_gate.weight | 0xb521a500 | 0x3d40000 | +| 100 | blk.10.ffn_norm.weight | 0xb8f5a500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0xb8f5f500 | 0x3d40000 | +| 102 | blk.11.attn_k.weight | 0xbcc9f500 | 0x1ea000 | +| 103 | blk.11.attn_norm.weight | 0xbce89500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0xbce8e500 | 0xb40000 | +| 105 | blk.11.attn_q.weight | 0xbd9ce500 | 0x7a8000 | +| 106 | blk.11.attn_v.weight | 0xbe176500 | 0x226000 | +| 107 | blk.11.ffn_down.weight | 0xbe39c500 | 0x44c0000 | +| 108 | blk.11.ffn_gate.weight | 0xc285c500 | 0x3d40000 | +| 109 | blk.11.ffn_norm.weight | 0xc659c500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0xc65a1500 | 0x3d40000 | +| 111 | blk.12.attn_k.weight | 0xca2e1500 | 0x1ea000 | +| 112 | blk.12.attn_norm.weight | 0xca4cb500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0xca4d0500 | 0xb40000 | +| 114 | blk.12.attn_q.weight | 0xcb010500 | 0x7a8000 | +| 115 | blk.12.attn_v.weight | 0xcb7b8500 | 0x226000 | +| 116 | blk.12.ffn_down.weight | 0xcb9de500 | 0x44c0000 | +| 117 | blk.12.ffn_gate.weight | 0xcfe9e500 | 0x3d40000 | +| 118 | blk.12.ffn_norm.weight | 0xd3bde500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0xd3be3500 | 0x3d40000 | +| 120 | blk.13.attn_k.weight | 0xd7923500 | 0x1ea000 | +| 121 | blk.13.attn_norm.weight | 0xd7b0d500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0xd7b12500 | 0xb40000 | +| 123 | blk.13.attn_q.weight | 0xd8652500 | 0x7a8000 | +| 124 | blk.13.attn_v.weight | 0xd8dfa500 | 0x226000 | +| 125 | blk.13.ffn_down.weight | 0xd9020500 | 0x44c0000 | +| 126 | blk.13.ffn_gate.weight | 0xdd4e0500 | 0x3d40000 | +| 127 | blk.13.ffn_norm.weight | 0xe1220500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0xe1225500 | 0x3d40000 | +| 129 | blk.14.attn_k.weight | 0xe4f65500 | 0x1ea000 | +| 130 | blk.14.attn_norm.weight | 0xe514f500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0xe5154500 | 0xb40000 | +| 132 | blk.14.attn_q.weight | 0xe5c94500 | 0x7a8000 | +| 133 | blk.14.attn_v.weight | 0xe643c500 | 0x226000 | +| 134 | blk.14.ffn_down.weight | 0xe6662500 | 0x44c0000 | +| 135 | blk.14.ffn_gate.weight | 0xeab22500 | 0x3d40000 | +| 136 | blk.14.ffn_norm.weight | 0xee862500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0xee867500 | 0x3d40000 | +| 138 | blk.15.attn_k.weight | 0xf25a7500 | 0x1ea000 | +| 139 | blk.15.attn_norm.weight | 0xf2791500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0xf2796500 | 0xb40000 | +| 141 | blk.15.attn_q.weight | 0xf32d6500 | 0x7a8000 | +| 142 | blk.15.attn_v.weight | 0xf3a7e500 | 0x226000 | +| 143 | blk.15.ffn_down.weight | 0xf3ca4500 | 0x44c0000 | +| 144 | blk.15.ffn_gate.weight | 0xf8164500 | 0x3d40000 | +| 145 | blk.15.ffn_norm.weight | 0xfbea4500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0xfbea9500 | 0x3d40000 | +| 147 | blk.16.attn_k.weight | 0xffbe9500 | 0x1ea000 | +| 148 | blk.16.attn_norm.weight | 0xffdd3500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0xffdd8500 | 0xb40000 | +| 150 | blk.16.attn_q.weight | 0x100918500 | 0x7a8000 | +| 151 | blk.16.attn_v.weight | 0x1010c0500 | 0x226000 | +| 152 | blk.16.ffn_down.weight | 0x1012e6500 | 0x44c0000 | +| 153 | blk.16.ffn_gate.weight | 0x1057a6500 | 0x3d40000 | +| 154 | blk.16.ffn_norm.weight | 0x1094e6500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0x1094eb500 | 0x3d40000 | +| 156 | blk.17.attn_k.weight | 0x10d22b500 | 0x226000 | +| 157 | blk.17.attn_norm.weight | 0x10d451500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0x10d456500 | 0xb40000 | +| 159 | blk.17.attn_q.weight | 0x10df96500 | 0x898000 | +| 160 | blk.17.attn_v.weight | 0x10e82e500 | 0x2d0000 | +| 161 | blk.17.ffn_down.weight | 0x10eafe500 | 0x44c0000 | +| 162 | blk.17.ffn_gate.weight | 0x112fbe500 | 0x3d40000 | +| 163 | blk.17.ffn_norm.weight | 0x116cfe500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0x116d03500 | 0x3d40000 | +| 165 | blk.18.attn_k.weight | 0x11aa43500 | 0x226000 | +| 166 | blk.18.attn_norm.weight | 0x11ac69500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0x11ac6e500 | 0xb40000 | +| 168 | blk.18.attn_q.weight | 0x11b7ae500 | 0x898000 | +| 169 | blk.18.attn_v.weight | 0x11c046500 | 0x2d0000 | +| 170 | blk.18.ffn_down.weight | 0x11c316500 | 0x44c0000 | +| 171 | blk.18.ffn_gate.weight | 0x1207d6500 | 0x3d40000 | +| 172 | blk.18.ffn_norm.weight | 0x124516500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x12451b500 | 0x3d40000 | +| 174 | blk.19.attn_k.weight | 0x12825b500 | 0x1ea000 | +| 175 | blk.19.attn_norm.weight | 0x128445500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x12844a500 | 0xb40000 | +| 177 | blk.19.attn_q.weight | 0x128f8a500 | 0x7a8000 | +| 178 | blk.19.attn_v.weight | 0x129732500 | 0x226000 | +| 179 | blk.19.ffn_down.weight | 0x129958500 | 0x44c0000 | +| 180 | blk.19.ffn_gate.weight | 0x12de18500 | 0x3d40000 | +| 181 | blk.19.ffn_norm.weight | 0x131b58500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x131b5d500 | 0x3d40000 | +| 183 | blk.20.attn_k.weight | 0x13589d500 | 0x226000 | +| 184 | blk.20.attn_norm.weight | 0x135ac3500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x135ac8500 | 0xb40000 | +| 186 | blk.20.attn_q.weight | 0x136608500 | 0x898000 | +| 187 | blk.20.attn_v.weight | 0x136ea0500 | 0x2d0000 | +| 188 | blk.20.ffn_down.weight | 0x137170500 | 0x44c0000 | +| 189 | blk.20.ffn_gate.weight | 0x13b630500 | 0x44c0000 | +| 190 | blk.20.ffn_norm.weight | 0x13faf0500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x13faf5500 | 0x44c0000 | +| 192 | blk.21.attn_k.weight | 0x143fb5500 | 0x1ea000 | +| 193 | blk.21.attn_norm.weight | 0x14419f500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x1441a4500 | 0xb40000 | +| 195 | blk.21.attn_q.weight | 0x144ce4500 | 0x7a8000 | +| 196 | blk.21.attn_v.weight | 0x14548c500 | 0x226000 | +| 197 | blk.21.ffn_down.weight | 0x1456b2500 | 0x44c0000 | +| 198 | blk.21.ffn_gate.weight | 0x149b72500 | 0x44c0000 | +| 199 | blk.21.ffn_norm.weight | 0x14e032500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x14e037500 | 0x44c0000 | +| 201 | blk.22.attn_k.weight | 0x1524f7500 | 0x226000 | +| 202 | blk.22.attn_norm.weight | 0x15271d500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x152722500 | 0xb40000 | +| 204 | blk.22.attn_q.weight | 0x153262500 | 0x898000 | +| 205 | blk.22.attn_v.weight | 0x153afa500 | 0x2d0000 | +| 206 | blk.22.ffn_down.weight | 0x153dca500 | 0x44c0000 | +| 207 | blk.22.ffn_gate.weight | 0x15828a500 | 0x44c0000 | +| 208 | blk.22.ffn_norm.weight | 0x15c74a500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x15c74f500 | 0x44c0000 | +| 210 | blk.23.attn_k.weight | 0x160c0f500 | 0x226000 | +| 211 | blk.23.attn_norm.weight | 0x160e35500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x160e3a500 | 0xb40000 | +| 213 | blk.23.attn_q.weight | 0x16197a500 | 0x898000 | +| 214 | blk.23.attn_v.weight | 0x162212500 | 0x2d0000 | +| 215 | blk.23.ffn_down.weight | 0x1624e2500 | 0x44c0000 | +| 216 | blk.23.ffn_gate.weight | 0x1669a2500 | 0x44c0000 | +| 217 | blk.23.ffn_norm.weight | 0x16ae62500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x16ae67500 | 0x44c0000 | +| 219 | blk.24.attn_k.weight | 0x16f327500 | 0x226000 | +| 220 | blk.24.attn_norm.weight | 0x16f54d500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x16f552500 | 0xb40000 | +| 222 | blk.24.attn_q.weight | 0x170092500 | 0x898000 | +| 223 | blk.24.attn_v.weight | 0x17092a500 | 0x2d0000 | +| 224 | blk.24.ffn_down.weight | 0x170bfa500 | 0x44c0000 | +| 225 | blk.24.ffn_gate.weight | 0x1750ba500 | 0x44c0000 | +| 226 | blk.24.ffn_norm.weight | 0x17957a500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x17957f500 | 0x44c0000 | +| 228 | blk.25.attn_k.weight | 0x17da3f500 | 0x226000 | +| 229 | blk.25.attn_norm.weight | 0x17dc65500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x17dc6a500 | 0xb40000 | +| 231 | blk.25.attn_q.weight | 0x17e7aa500 | 0x898000 | +| 232 | blk.25.attn_v.weight | 0x17f042500 | 0x2d0000 | +| 233 | blk.25.ffn_down.weight | 0x17f312500 | 0x44c0000 | +| 234 | blk.25.ffn_gate.weight | 0x1837d2500 | 0x44c0000 | +| 235 | blk.25.ffn_norm.weight | 0x187c92500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x187c97500 | 0x44c0000 | +| 237 | blk.26.attn_k.weight | 0x18c157500 | 0x226000 | +| 238 | blk.26.attn_norm.weight | 0x18c37d500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x18c382500 | 0xb40000 | +| 240 | blk.26.attn_q.weight | 0x18cec2500 | 0x898000 | +| 241 | blk.26.attn_v.weight | 0x18d75a500 | 0x2d0000 | +| 242 | blk.26.ffn_down.weight | 0x18da2a500 | 0x44c0000 | +| 243 | blk.26.ffn_gate.weight | 0x191eea500 | 0x44c0000 | +| 244 | blk.26.ffn_norm.weight | 0x1963aa500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x1963af500 | 0x44c0000 | +| 246 | blk.27.attn_k.weight | 0x19a86f500 | 0x1ea000 | +| 247 | blk.27.attn_norm.weight | 0x19aa59500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x19aa5e500 | 0xb40000 | +| 249 | blk.27.attn_q.weight | 0x19b59e500 | 0x7a8000 | +| 250 | blk.27.attn_v.weight | 0x19bd46500 | 0x226000 | +| 251 | blk.27.ffn_down.weight | 0x19bf6c500 | 0x44c0000 | +| 252 | blk.27.ffn_gate.weight | 0x1a042c500 | 0x44c0000 | +| 253 | blk.27.ffn_norm.weight | 0x1a48ec500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x1a48f1500 | 0x44c0000 | +| 255 | blk.28.attn_k.weight | 0x1a8db1500 | 0x226000 | +| 256 | blk.28.attn_norm.weight | 0x1a8fd7500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x1a8fdc500 | 0xb40000 | +| 258 | blk.28.attn_q.weight | 0x1a9b1c500 | 0x898000 | +| 259 | blk.28.attn_v.weight | 0x1aa3b4500 | 0x2d0000 | +| 260 | blk.28.ffn_down.weight | 0x1aa684500 | 0x44c0000 | +| 261 | blk.28.ffn_gate.weight | 0x1aeb44500 | 0x44c0000 | +| 262 | blk.28.ffn_norm.weight | 0x1b3004500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x1b3009500 | 0x44c0000 | +| 264 | blk.29.attn_k.weight | 0x1b74c9500 | 0x226000 | +| 265 | blk.29.attn_norm.weight | 0x1b76ef500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x1b76f4500 | 0xb40000 | +| 267 | blk.29.attn_q.weight | 0x1b8234500 | 0x898000 | +| 268 | blk.29.attn_v.weight | 0x1b8acc500 | 0x2d0000 | +| 269 | blk.29.ffn_down.weight | 0x1b8d9c500 | 0x44c0000 | +| 270 | blk.29.ffn_gate.weight | 0x1bd25c500 | 0x44c0000 | +| 271 | blk.29.ffn_norm.weight | 0x1c171c500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x1c1721500 | 0x44c0000 | +| 273 | blk.30.attn_k.weight | 0x1c5be1500 | 0x226000 | +| 274 | blk.30.attn_norm.weight | 0x1c5e07500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x1c5e0c500 | 0xb40000 | +| 276 | blk.30.attn_q.weight | 0x1c694c500 | 0x898000 | +| 277 | blk.30.attn_v.weight | 0x1c71e4500 | 0x2d0000 | +| 278 | blk.30.ffn_down.weight | 0x1c74b4500 | 0x44c0000 | +| 279 | blk.30.ffn_gate.weight | 0x1cb974500 | 0x44c0000 | +| 280 | blk.30.ffn_norm.weight | 0x1cfe34500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x1cfe39500 | 0x44c0000 | +| 282 | blk.31.attn_k.weight | 0x1d42f9500 | 0x226000 | +| 283 | blk.31.attn_norm.weight | 0x1d451f500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x1d4524500 | 0xb40000 | +| 285 | blk.31.attn_q.weight | 0x1d5064500 | 0x898000 | +| 286 | blk.31.attn_v.weight | 0x1d58fc500 | 0x2d0000 | +| 287 | blk.31.ffn_down.weight | 0x1d5bcc500 | 0x44c0000 | +| 288 | blk.31.ffn_gate.weight | 0x1da08c500 | 0x44c0000 | +| 289 | blk.31.ffn_norm.weight | 0x1de54c500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x1de551500 | 0x44c0000 | +| 291 | blk.32.attn_k.weight | 0x1e2a11500 | 0x226000 | +| 292 | blk.32.attn_norm.weight | 0x1e2c37500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x1e2c3c500 | 0xb40000 | +| 294 | blk.32.attn_q.weight | 0x1e377c500 | 0x898000 | +| 295 | blk.32.attn_v.weight | 0x1e4014500 | 0x2d0000 | +| 296 | blk.32.ffn_down.weight | 0x1e42e4500 | 0x44c0000 | +| 297 | blk.32.ffn_gate.weight | 0x1e87a4500 | 0x44c0000 | +| 298 | blk.32.ffn_norm.weight | 0x1ecc64500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x1ecc69500 | 0x44c0000 | +| 300 | blk.33.attn_k.weight | 0x1f1129500 | 0x226000 | +| 301 | blk.33.attn_norm.weight | 0x1f134f500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x1f1354500 | 0xb40000 | +| 303 | blk.33.attn_q.weight | 0x1f1e94500 | 0x898000 | +| 304 | blk.33.attn_v.weight | 0x1f272c500 | 0x2d0000 | +| 305 | blk.33.ffn_down.weight | 0x1f29fc500 | 0x44c0000 | +| 306 | blk.33.ffn_gate.weight | 0x1f6ebc500 | 0x44c0000 | +| 307 | blk.33.ffn_norm.weight | 0x1fb37c500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x1fb381500 | 0x44c0000 | +| 309 | blk.34.attn_k.weight | 0x1ff841500 | 0x226000 | +| 310 | blk.34.attn_norm.weight | 0x1ffa67500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x1ffa6c500 | 0xb40000 | +| 312 | blk.34.attn_q.weight | 0x2005ac500 | 0x898000 | +| 313 | blk.34.attn_v.weight | 0x200e44500 | 0x2d0000 | +| 314 | blk.34.ffn_down.weight | 0x201114500 | 0x44c0000 | +| 315 | blk.34.ffn_gate.weight | 0x2055d4500 | 0x44c0000 | +| 316 | blk.34.ffn_norm.weight | 0x209a94500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x209a99500 | 0x44c0000 | +| 318 | blk.35.attn_k.weight | 0x20df59500 | 0x226000 | +| 319 | blk.35.attn_norm.weight | 0x20e17f500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x20e184500 | 0xb40000 | +| 321 | blk.35.attn_q.weight | 0x20ecc4500 | 0x898000 | +| 322 | blk.35.attn_v.weight | 0x20f55c500 | 0x2d0000 | +| 323 | blk.35.ffn_down.weight | 0x20f82c500 | 0x44c0000 | +| 324 | blk.35.ffn_gate.weight | 0x213cec500 | 0x44c0000 | +| 325 | blk.35.ffn_norm.weight | 0x2181ac500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x2181b1500 | 0x44c0000 | +| 327 | blk.36.attn_k.weight | 0x21c671500 | 0x226000 | +| 328 | blk.36.attn_norm.weight | 0x21c897500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x21c89c500 | 0xb40000 | +| 330 | blk.36.attn_q.weight | 0x21d3dc500 | 0x898000 | +| 331 | blk.36.attn_v.weight | 0x21dc74500 | 0x2d0000 | +| 332 | blk.36.ffn_down.weight | 0x21df44500 | 0x44c0000 | +| 333 | blk.36.ffn_gate.weight | 0x222404500 | 0x44c0000 | +| 334 | blk.36.ffn_norm.weight | 0x2268c4500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x2268c9500 | 0x44c0000 | +| 336 | blk.37.attn_k.weight | 0x22ad89500 | 0x226000 | +| 337 | blk.37.attn_norm.weight | 0x22afaf500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x22afb4500 | 0xb40000 | +| 339 | blk.37.attn_q.weight | 0x22baf4500 | 0x898000 | +| 340 | blk.37.attn_v.weight | 0x22c38c500 | 0x2d0000 | +| 341 | blk.37.ffn_down.weight | 0x22c65c500 | 0x44c0000 | +| 342 | blk.37.ffn_gate.weight | 0x230b1c500 | 0x44c0000 | +| 343 | blk.37.ffn_norm.weight | 0x234fdc500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x234fe1500 | 0x44c0000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:------| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | IQ3_S | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | IQ3_S | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-IQ3_S.md b/scores/Dolphin-Mistral-24B-Venice-Edition-IQ3_S.md new file mode 100644 index 0000000..7aa10a9 --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-IQ3_S.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-IQ3_S.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 26 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-IQ3\_S.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-iq3_sgguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0xf500000 | +| 1 | output_norm.weight | 0xfc84500 | 0x5000 | +| 2 | token_embd.weight | 0xfc89500 | 0xf500000 | +| 3 | blk.0.attn_k.weight | 0x1f189500 | 0x1ea000 | +| 4 | blk.0.attn_norm.weight | 0x1f373500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x1f378500 | 0x898000 | +| 6 | blk.0.attn_q.weight | 0x1fc10500 | 0x7a8000 | +| 7 | blk.0.attn_v.weight | 0x203b8500 | 0x226000 | +| 8 | blk.0.ffn_down.weight | 0x205de500 | 0x44c0000 | +| 9 | blk.0.ffn_gate.weight | 0x24a9e500 | 0x3d40000 | +| 10 | blk.0.ffn_norm.weight | 0x287de500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x287e3500 | 0x3d40000 | +| 12 | blk.1.attn_k.weight | 0x2c523500 | 0x1ea000 | +| 13 | blk.1.attn_norm.weight | 0x2c70d500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x2c712500 | 0x898000 | +| 15 | blk.1.attn_q.weight | 0x2cfaa500 | 0x7a8000 | +| 16 | blk.1.attn_v.weight | 0x2d752500 | 0x226000 | +| 17 | blk.1.ffn_down.weight | 0x2d978500 | 0x44c0000 | +| 18 | blk.1.ffn_gate.weight | 0x31e38500 | 0x3d40000 | +| 19 | blk.1.ffn_norm.weight | 0x35b78500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x35b7d500 | 0x3d40000 | +| 21 | blk.2.attn_k.weight | 0x398bd500 | 0x1ea000 | +| 22 | blk.2.attn_norm.weight | 0x39aa7500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x39aac500 | 0x898000 | +| 24 | blk.2.attn_q.weight | 0x3a344500 | 0x7a8000 | +| 25 | blk.2.attn_v.weight | 0x3aaec500 | 0x226000 | +| 26 | blk.2.ffn_down.weight | 0x3ad12500 | 0x44c0000 | +| 27 | blk.2.ffn_gate.weight | 0x3f1d2500 | 0x3d40000 | +| 28 | blk.2.ffn_norm.weight | 0x42f12500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x42f17500 | 0x3d40000 | +| 30 | blk.3.attn_k.weight | 0x46c57500 | 0x1ea000 | +| 31 | blk.3.attn_norm.weight | 0x46e41500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x46e46500 | 0x898000 | +| 33 | blk.3.attn_q.weight | 0x476de500 | 0x7a8000 | +| 34 | blk.3.attn_v.weight | 0x47e86500 | 0x226000 | +| 35 | blk.3.ffn_down.weight | 0x480ac500 | 0x44c0000 | +| 36 | blk.3.ffn_gate.weight | 0x4c56c500 | 0x3d40000 | +| 37 | blk.3.ffn_norm.weight | 0x502ac500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0x502b1500 | 0x3d40000 | +| 39 | blk.4.attn_k.weight | 0x53ff1500 | 0x1ea000 | +| 40 | blk.4.attn_norm.weight | 0x541db500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0x541e0500 | 0x898000 | +| 42 | blk.4.attn_q.weight | 0x54a78500 | 0x7a8000 | +| 43 | blk.4.attn_v.weight | 0x55220500 | 0x226000 | +| 44 | blk.4.ffn_down.weight | 0x55446500 | 0x44c0000 | +| 45 | blk.4.ffn_gate.weight | 0x59906500 | 0x3d40000 | +| 46 | blk.4.ffn_norm.weight | 0x5d646500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0x5d64b500 | 0x3d40000 | +| 48 | blk.5.attn_k.weight | 0x6138b500 | 0x1ea000 | +| 49 | blk.5.attn_norm.weight | 0x61575500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0x6157a500 | 0x898000 | +| 51 | blk.5.attn_q.weight | 0x61e12500 | 0x7a8000 | +| 52 | blk.5.attn_v.weight | 0x625ba500 | 0x226000 | +| 53 | blk.5.ffn_down.weight | 0x627e0500 | 0x44c0000 | +| 54 | blk.5.ffn_gate.weight | 0x66ca0500 | 0x3d40000 | +| 55 | blk.5.ffn_norm.weight | 0x6a9e0500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0x6a9e5500 | 0x3d40000 | +| 57 | blk.6.attn_k.weight | 0x6e725500 | 0x1ea000 | +| 58 | blk.6.attn_norm.weight | 0x6e90f500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0x6e914500 | 0x898000 | +| 60 | blk.6.attn_q.weight | 0x6f1ac500 | 0x7a8000 | +| 61 | blk.6.attn_v.weight | 0x6f954500 | 0x226000 | +| 62 | blk.6.ffn_down.weight | 0x6fb7a500 | 0x44c0000 | +| 63 | blk.6.ffn_gate.weight | 0x7403a500 | 0x3d40000 | +| 64 | blk.6.ffn_norm.weight | 0x77d7a500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0x77d7f500 | 0x3d40000 | +| 66 | blk.7.attn_k.weight | 0x7babf500 | 0x1ea000 | +| 67 | blk.7.attn_norm.weight | 0x7bca9500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0x7bcae500 | 0x898000 | +| 69 | blk.7.attn_q.weight | 0x7c546500 | 0x7a8000 | +| 70 | blk.7.attn_v.weight | 0x7ccee500 | 0x226000 | +| 71 | blk.7.ffn_down.weight | 0x7cf14500 | 0x44c0000 | +| 72 | blk.7.ffn_gate.weight | 0x813d4500 | 0x3d40000 | +| 73 | blk.7.ffn_norm.weight | 0x85114500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0x85119500 | 0x3d40000 | +| 75 | blk.8.attn_k.weight | 0x88e59500 | 0x1ea000 | +| 76 | blk.8.attn_norm.weight | 0x89043500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0x89048500 | 0x898000 | +| 78 | blk.8.attn_q.weight | 0x898e0500 | 0x7a8000 | +| 79 | blk.8.attn_v.weight | 0x8a088500 | 0x226000 | +| 80 | blk.8.ffn_down.weight | 0x8a2ae500 | 0x44c0000 | +| 81 | blk.8.ffn_gate.weight | 0x8e76e500 | 0x3d40000 | +| 82 | blk.8.ffn_norm.weight | 0x924ae500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0x924b3500 | 0x3d40000 | +| 84 | blk.9.attn_k.weight | 0x961f3500 | 0x1ea000 | +| 85 | blk.9.attn_norm.weight | 0x963dd500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0x963e2500 | 0x898000 | +| 87 | blk.9.attn_q.weight | 0x96c7a500 | 0x7a8000 | +| 88 | blk.9.attn_v.weight | 0x97422500 | 0x226000 | +| 89 | blk.9.ffn_down.weight | 0x97648500 | 0x44c0000 | +| 90 | blk.9.ffn_gate.weight | 0x9bb08500 | 0x3d40000 | +| 91 | blk.9.ffn_norm.weight | 0x9f848500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0x9f84d500 | 0x3d40000 | +| 93 | blk.10.attn_k.weight | 0xa358d500 | 0x1ea000 | +| 94 | blk.10.attn_norm.weight | 0xa3777500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0xa377c500 | 0x898000 | +| 96 | blk.10.attn_q.weight | 0xa4014500 | 0x7a8000 | +| 97 | blk.10.attn_v.weight | 0xa47bc500 | 0x226000 | +| 98 | blk.10.ffn_down.weight | 0xa49e2500 | 0x44c0000 | +| 99 | blk.10.ffn_gate.weight | 0xa8ea2500 | 0x3d40000 | +| 100 | blk.10.ffn_norm.weight | 0xacbe2500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0xacbe7500 | 0x3d40000 | +| 102 | blk.11.attn_k.weight | 0xb0927500 | 0x1ea000 | +| 103 | blk.11.attn_norm.weight | 0xb0b11500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0xb0b16500 | 0x898000 | +| 105 | blk.11.attn_q.weight | 0xb13ae500 | 0x7a8000 | +| 106 | blk.11.attn_v.weight | 0xb1b56500 | 0x226000 | +| 107 | blk.11.ffn_down.weight | 0xb1d7c500 | 0x44c0000 | +| 108 | blk.11.ffn_gate.weight | 0xb623c500 | 0x3d40000 | +| 109 | blk.11.ffn_norm.weight | 0xb9f7c500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0xb9f81500 | 0x3d40000 | +| 111 | blk.12.attn_k.weight | 0xbdcc1500 | 0x1ea000 | +| 112 | blk.12.attn_norm.weight | 0xbdeab500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0xbdeb0500 | 0x898000 | +| 114 | blk.12.attn_q.weight | 0xbe748500 | 0x7a8000 | +| 115 | blk.12.attn_v.weight | 0xbeef0500 | 0x226000 | +| 116 | blk.12.ffn_down.weight | 0xbf116500 | 0x44c0000 | +| 117 | blk.12.ffn_gate.weight | 0xc35d6500 | 0x3d40000 | +| 118 | blk.12.ffn_norm.weight | 0xc7316500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0xc731b500 | 0x3d40000 | +| 120 | blk.13.attn_k.weight | 0xcb05b500 | 0x1ea000 | +| 121 | blk.13.attn_norm.weight | 0xcb245500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0xcb24a500 | 0x898000 | +| 123 | blk.13.attn_q.weight | 0xcbae2500 | 0x7a8000 | +| 124 | blk.13.attn_v.weight | 0xcc28a500 | 0x226000 | +| 125 | blk.13.ffn_down.weight | 0xcc4b0500 | 0x44c0000 | +| 126 | blk.13.ffn_gate.weight | 0xd0970500 | 0x3d40000 | +| 127 | blk.13.ffn_norm.weight | 0xd46b0500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0xd46b5500 | 0x3d40000 | +| 129 | blk.14.attn_k.weight | 0xd83f5500 | 0x1ea000 | +| 130 | blk.14.attn_norm.weight | 0xd85df500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0xd85e4500 | 0x898000 | +| 132 | blk.14.attn_q.weight | 0xd8e7c500 | 0x7a8000 | +| 133 | blk.14.attn_v.weight | 0xd9624500 | 0x226000 | +| 134 | blk.14.ffn_down.weight | 0xd984a500 | 0x44c0000 | +| 135 | blk.14.ffn_gate.weight | 0xddd0a500 | 0x3d40000 | +| 136 | blk.14.ffn_norm.weight | 0xe1a4a500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0xe1a4f500 | 0x3d40000 | +| 138 | blk.15.attn_k.weight | 0xe578f500 | 0x1ea000 | +| 139 | blk.15.attn_norm.weight | 0xe5979500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0xe597e500 | 0x898000 | +| 141 | blk.15.attn_q.weight | 0xe6216500 | 0x7a8000 | +| 142 | blk.15.attn_v.weight | 0xe69be500 | 0x226000 | +| 143 | blk.15.ffn_down.weight | 0xe6be4500 | 0x44c0000 | +| 144 | blk.15.ffn_gate.weight | 0xeb0a4500 | 0x3d40000 | +| 145 | blk.15.ffn_norm.weight | 0xeede4500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0xeede9500 | 0x3d40000 | +| 147 | blk.16.attn_k.weight | 0xf2b29500 | 0x1ea000 | +| 148 | blk.16.attn_norm.weight | 0xf2d13500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0xf2d18500 | 0x898000 | +| 150 | blk.16.attn_q.weight | 0xf35b0500 | 0x7a8000 | +| 151 | blk.16.attn_v.weight | 0xf3d58500 | 0x226000 | +| 152 | blk.16.ffn_down.weight | 0xf3f7e500 | 0x44c0000 | +| 153 | blk.16.ffn_gate.weight | 0xf843e500 | 0x3d40000 | +| 154 | blk.16.ffn_norm.weight | 0xfc17e500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0xfc183500 | 0x3d40000 | +| 156 | blk.17.attn_k.weight | 0xffec3500 | 0x226000 | +| 157 | blk.17.attn_norm.weight | 0x1000e9500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0x1000ee500 | 0x898000 | +| 159 | blk.17.attn_q.weight | 0x100986500 | 0x898000 | +| 160 | blk.17.attn_v.weight | 0x10121e500 | 0x226000 | +| 161 | blk.17.ffn_down.weight | 0x101444500 | 0x44c0000 | +| 162 | blk.17.ffn_gate.weight | 0x105904500 | 0x3d40000 | +| 163 | blk.17.ffn_norm.weight | 0x109644500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0x109649500 | 0x3d40000 | +| 165 | blk.18.attn_k.weight | 0x10d389500 | 0x226000 | +| 166 | blk.18.attn_norm.weight | 0x10d5af500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0x10d5b4500 | 0x898000 | +| 168 | blk.18.attn_q.weight | 0x10de4c500 | 0x898000 | +| 169 | blk.18.attn_v.weight | 0x10e6e4500 | 0x226000 | +| 170 | blk.18.ffn_down.weight | 0x10e90a500 | 0x44c0000 | +| 171 | blk.18.ffn_gate.weight | 0x112dca500 | 0x3d40000 | +| 172 | blk.18.ffn_norm.weight | 0x116b0a500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x116b0f500 | 0x3d40000 | +| 174 | blk.19.attn_k.weight | 0x11a84f500 | 0x1ea000 | +| 175 | blk.19.attn_norm.weight | 0x11aa39500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x11aa3e500 | 0x898000 | +| 177 | blk.19.attn_q.weight | 0x11b2d6500 | 0x7a8000 | +| 178 | blk.19.attn_v.weight | 0x11ba7e500 | 0x226000 | +| 179 | blk.19.ffn_down.weight | 0x11bca4500 | 0x44c0000 | +| 180 | blk.19.ffn_gate.weight | 0x120164500 | 0x3d40000 | +| 181 | blk.19.ffn_norm.weight | 0x123ea4500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x123ea9500 | 0x3d40000 | +| 183 | blk.20.attn_k.weight | 0x127be9500 | 0x226000 | +| 184 | blk.20.attn_norm.weight | 0x127e0f500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x127e14500 | 0x898000 | +| 186 | blk.20.attn_q.weight | 0x1286ac500 | 0x898000 | +| 187 | blk.20.attn_v.weight | 0x128f44500 | 0x226000 | +| 188 | blk.20.ffn_down.weight | 0x12916a500 | 0x44c0000 | +| 189 | blk.20.ffn_gate.weight | 0x12d62a500 | 0x44c0000 | +| 190 | blk.20.ffn_norm.weight | 0x131aea500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x131aef500 | 0x44c0000 | +| 192 | blk.21.attn_k.weight | 0x135faf500 | 0x1ea000 | +| 193 | blk.21.attn_norm.weight | 0x136199500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x13619e500 | 0x898000 | +| 195 | blk.21.attn_q.weight | 0x136a36500 | 0x7a8000 | +| 196 | blk.21.attn_v.weight | 0x1371de500 | 0x226000 | +| 197 | blk.21.ffn_down.weight | 0x137404500 | 0x44c0000 | +| 198 | blk.21.ffn_gate.weight | 0x13b8c4500 | 0x44c0000 | +| 199 | blk.21.ffn_norm.weight | 0x13fd84500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x13fd89500 | 0x44c0000 | +| 201 | blk.22.attn_k.weight | 0x144249500 | 0x226000 | +| 202 | blk.22.attn_norm.weight | 0x14446f500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x144474500 | 0x898000 | +| 204 | blk.22.attn_q.weight | 0x144d0c500 | 0x898000 | +| 205 | blk.22.attn_v.weight | 0x1455a4500 | 0x226000 | +| 206 | blk.22.ffn_down.weight | 0x1457ca500 | 0x44c0000 | +| 207 | blk.22.ffn_gate.weight | 0x149c8a500 | 0x44c0000 | +| 208 | blk.22.ffn_norm.weight | 0x14e14a500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x14e14f500 | 0x44c0000 | +| 210 | blk.23.attn_k.weight | 0x15260f500 | 0x226000 | +| 211 | blk.23.attn_norm.weight | 0x152835500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x15283a500 | 0x898000 | +| 213 | blk.23.attn_q.weight | 0x1530d2500 | 0x898000 | +| 214 | blk.23.attn_v.weight | 0x15396a500 | 0x226000 | +| 215 | blk.23.ffn_down.weight | 0x153b90500 | 0x44c0000 | +| 216 | blk.23.ffn_gate.weight | 0x158050500 | 0x44c0000 | +| 217 | blk.23.ffn_norm.weight | 0x15c510500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x15c515500 | 0x44c0000 | +| 219 | blk.24.attn_k.weight | 0x1609d5500 | 0x226000 | +| 220 | blk.24.attn_norm.weight | 0x160bfb500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x160c00500 | 0x898000 | +| 222 | blk.24.attn_q.weight | 0x161498500 | 0x898000 | +| 223 | blk.24.attn_v.weight | 0x161d30500 | 0x226000 | +| 224 | blk.24.ffn_down.weight | 0x161f56500 | 0x44c0000 | +| 225 | blk.24.ffn_gate.weight | 0x166416500 | 0x44c0000 | +| 226 | blk.24.ffn_norm.weight | 0x16a8d6500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x16a8db500 | 0x44c0000 | +| 228 | blk.25.attn_k.weight | 0x16ed9b500 | 0x226000 | +| 229 | blk.25.attn_norm.weight | 0x16efc1500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x16efc6500 | 0x898000 | +| 231 | blk.25.attn_q.weight | 0x16f85e500 | 0x898000 | +| 232 | blk.25.attn_v.weight | 0x1700f6500 | 0x226000 | +| 233 | blk.25.ffn_down.weight | 0x17031c500 | 0x44c0000 | +| 234 | blk.25.ffn_gate.weight | 0x1747dc500 | 0x44c0000 | +| 235 | blk.25.ffn_norm.weight | 0x178c9c500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x178ca1500 | 0x44c0000 | +| 237 | blk.26.attn_k.weight | 0x17d161500 | 0x226000 | +| 238 | blk.26.attn_norm.weight | 0x17d387500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x17d38c500 | 0x898000 | +| 240 | blk.26.attn_q.weight | 0x17dc24500 | 0x898000 | +| 241 | blk.26.attn_v.weight | 0x17e4bc500 | 0x226000 | +| 242 | blk.26.ffn_down.weight | 0x17e6e2500 | 0x44c0000 | +| 243 | blk.26.ffn_gate.weight | 0x182ba2500 | 0x44c0000 | +| 244 | blk.26.ffn_norm.weight | 0x187062500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x187067500 | 0x44c0000 | +| 246 | blk.27.attn_k.weight | 0x18b527500 | 0x1ea000 | +| 247 | blk.27.attn_norm.weight | 0x18b711500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x18b716500 | 0x898000 | +| 249 | blk.27.attn_q.weight | 0x18bfae500 | 0x7a8000 | +| 250 | blk.27.attn_v.weight | 0x18c756500 | 0x226000 | +| 251 | blk.27.ffn_down.weight | 0x18c97c500 | 0x44c0000 | +| 252 | blk.27.ffn_gate.weight | 0x190e3c500 | 0x44c0000 | +| 253 | blk.27.ffn_norm.weight | 0x1952fc500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x195301500 | 0x44c0000 | +| 255 | blk.28.attn_k.weight | 0x1997c1500 | 0x226000 | +| 256 | blk.28.attn_norm.weight | 0x1999e7500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x1999ec500 | 0x898000 | +| 258 | blk.28.attn_q.weight | 0x19a284500 | 0x898000 | +| 259 | blk.28.attn_v.weight | 0x19ab1c500 | 0x226000 | +| 260 | blk.28.ffn_down.weight | 0x19ad42500 | 0x44c0000 | +| 261 | blk.28.ffn_gate.weight | 0x19f202500 | 0x44c0000 | +| 262 | blk.28.ffn_norm.weight | 0x1a36c2500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x1a36c7500 | 0x44c0000 | +| 264 | blk.29.attn_k.weight | 0x1a7b87500 | 0x226000 | +| 265 | blk.29.attn_norm.weight | 0x1a7dad500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x1a7db2500 | 0x898000 | +| 267 | blk.29.attn_q.weight | 0x1a864a500 | 0x898000 | +| 268 | blk.29.attn_v.weight | 0x1a8ee2500 | 0x226000 | +| 269 | blk.29.ffn_down.weight | 0x1a9108500 | 0x44c0000 | +| 270 | blk.29.ffn_gate.weight | 0x1ad5c8500 | 0x44c0000 | +| 271 | blk.29.ffn_norm.weight | 0x1b1a88500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x1b1a8d500 | 0x44c0000 | +| 273 | blk.30.attn_k.weight | 0x1b5f4d500 | 0x226000 | +| 274 | blk.30.attn_norm.weight | 0x1b6173500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x1b6178500 | 0x898000 | +| 276 | blk.30.attn_q.weight | 0x1b6a10500 | 0x898000 | +| 277 | blk.30.attn_v.weight | 0x1b72a8500 | 0x226000 | +| 278 | blk.30.ffn_down.weight | 0x1b74ce500 | 0x44c0000 | +| 279 | blk.30.ffn_gate.weight | 0x1bb98e500 | 0x44c0000 | +| 280 | blk.30.ffn_norm.weight | 0x1bfe4e500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x1bfe53500 | 0x44c0000 | +| 282 | blk.31.attn_k.weight | 0x1c4313500 | 0x226000 | +| 283 | blk.31.attn_norm.weight | 0x1c4539500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x1c453e500 | 0x898000 | +| 285 | blk.31.attn_q.weight | 0x1c4dd6500 | 0x898000 | +| 286 | blk.31.attn_v.weight | 0x1c566e500 | 0x226000 | +| 287 | blk.31.ffn_down.weight | 0x1c5894500 | 0x44c0000 | +| 288 | blk.31.ffn_gate.weight | 0x1c9d54500 | 0x44c0000 | +| 289 | blk.31.ffn_norm.weight | 0x1ce214500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x1ce219500 | 0x44c0000 | +| 291 | blk.32.attn_k.weight | 0x1d26d9500 | 0x226000 | +| 292 | blk.32.attn_norm.weight | 0x1d28ff500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x1d2904500 | 0x898000 | +| 294 | blk.32.attn_q.weight | 0x1d319c500 | 0x898000 | +| 295 | blk.32.attn_v.weight | 0x1d3a34500 | 0x226000 | +| 296 | blk.32.ffn_down.weight | 0x1d3c5a500 | 0x44c0000 | +| 297 | blk.32.ffn_gate.weight | 0x1d811a500 | 0x44c0000 | +| 298 | blk.32.ffn_norm.weight | 0x1dc5da500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x1dc5df500 | 0x44c0000 | +| 300 | blk.33.attn_k.weight | 0x1e0a9f500 | 0x226000 | +| 301 | blk.33.attn_norm.weight | 0x1e0cc5500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x1e0cca500 | 0x898000 | +| 303 | blk.33.attn_q.weight | 0x1e1562500 | 0x898000 | +| 304 | blk.33.attn_v.weight | 0x1e1dfa500 | 0x226000 | +| 305 | blk.33.ffn_down.weight | 0x1e2020500 | 0x44c0000 | +| 306 | blk.33.ffn_gate.weight | 0x1e64e0500 | 0x44c0000 | +| 307 | blk.33.ffn_norm.weight | 0x1ea9a0500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x1ea9a5500 | 0x44c0000 | +| 309 | blk.34.attn_k.weight | 0x1eee65500 | 0x226000 | +| 310 | blk.34.attn_norm.weight | 0x1ef08b500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x1ef090500 | 0x898000 | +| 312 | blk.34.attn_q.weight | 0x1ef928500 | 0x898000 | +| 313 | blk.34.attn_v.weight | 0x1f01c0500 | 0x226000 | +| 314 | blk.34.ffn_down.weight | 0x1f03e6500 | 0x44c0000 | +| 315 | blk.34.ffn_gate.weight | 0x1f48a6500 | 0x44c0000 | +| 316 | blk.34.ffn_norm.weight | 0x1f8d66500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x1f8d6b500 | 0x44c0000 | +| 318 | blk.35.attn_k.weight | 0x1fd22b500 | 0x226000 | +| 319 | blk.35.attn_norm.weight | 0x1fd451500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x1fd456500 | 0x898000 | +| 321 | blk.35.attn_q.weight | 0x1fdcee500 | 0x898000 | +| 322 | blk.35.attn_v.weight | 0x1fe586500 | 0x226000 | +| 323 | blk.35.ffn_down.weight | 0x1fe7ac500 | 0x44c0000 | +| 324 | blk.35.ffn_gate.weight | 0x202c6c500 | 0x44c0000 | +| 325 | blk.35.ffn_norm.weight | 0x20712c500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x207131500 | 0x44c0000 | +| 327 | blk.36.attn_k.weight | 0x20b5f1500 | 0x226000 | +| 328 | blk.36.attn_norm.weight | 0x20b817500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x20b81c500 | 0x898000 | +| 330 | blk.36.attn_q.weight | 0x20c0b4500 | 0x898000 | +| 331 | blk.36.attn_v.weight | 0x20c94c500 | 0x226000 | +| 332 | blk.36.ffn_down.weight | 0x20cb72500 | 0x44c0000 | +| 333 | blk.36.ffn_gate.weight | 0x211032500 | 0x44c0000 | +| 334 | blk.36.ffn_norm.weight | 0x2154f2500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x2154f7500 | 0x44c0000 | +| 336 | blk.37.attn_k.weight | 0x2199b7500 | 0x226000 | +| 337 | blk.37.attn_norm.weight | 0x219bdd500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x219be2500 | 0x898000 | +| 339 | blk.37.attn_q.weight | 0x21a47a500 | 0x898000 | +| 340 | blk.37.attn_v.weight | 0x21ad12500 | 0x226000 | +| 341 | blk.37.ffn_down.weight | 0x21af38500 | 0x44c0000 | +| 342 | blk.37.ffn_gate.weight | 0x21f3f8500 | 0x44c0000 | +| 343 | blk.37.ffn_norm.weight | 0x2238b8500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x2238bd500 | 0x44c0000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:--------| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | IQ3_XXS | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | IQ3_XXS | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:--------| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_XXS | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:--------| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_XXS | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_XXS | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:------| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ3_S | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ3_S | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-IQ4_NL.md b/scores/Dolphin-Mistral-24B-Venice-Edition-IQ4_NL.md new file mode 100644 index 0000000..4513b65 --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-IQ4_NL.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-IQ4_NL.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 25 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-IQ4\_NL.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-iq4_nlgguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0x16800000 | +| 1 | output_norm.weight | 0x16f84500 | 0x5000 | +| 2 | token_embd.weight | 0x16f89500 | 0x11300000 | +| 3 | blk.0.attn_k.weight | 0x28289500 | 0x226000 | +| 4 | blk.0.attn_norm.weight | 0x284af500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x284b4500 | 0xb40000 | +| 6 | blk.0.attn_q.weight | 0x28ff4500 | 0x898000 | +| 7 | blk.0.attn_v.weight | 0x2988c500 | 0x2a8000 | +| 8 | blk.0.ffn_down.weight | 0x29b34500 | 0x5a00000 | +| 9 | blk.0.ffn_gate.weight | 0x2f534500 | 0x44c0000 | +| 10 | blk.0.ffn_norm.weight | 0x339f4500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x339f9500 | 0x44c0000 | +| 12 | blk.1.attn_k.weight | 0x37eb9500 | 0x226000 | +| 13 | blk.1.attn_norm.weight | 0x380df500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x380e4500 | 0xb40000 | +| 15 | blk.1.attn_q.weight | 0x38c24500 | 0x898000 | +| 16 | blk.1.attn_v.weight | 0x394bc500 | 0x2a8000 | +| 17 | blk.1.ffn_down.weight | 0x39764500 | 0x5a00000 | +| 18 | blk.1.ffn_gate.weight | 0x3f164500 | 0x44c0000 | +| 19 | blk.1.ffn_norm.weight | 0x43624500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x43629500 | 0x44c0000 | +| 21 | blk.2.attn_k.weight | 0x47ae9500 | 0x226000 | +| 22 | blk.2.attn_norm.weight | 0x47d0f500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x47d14500 | 0xb40000 | +| 24 | blk.2.attn_q.weight | 0x48854500 | 0x898000 | +| 25 | blk.2.attn_v.weight | 0x490ec500 | 0x2a8000 | +| 26 | blk.2.ffn_down.weight | 0x49394500 | 0x5a00000 | +| 27 | blk.2.ffn_gate.weight | 0x4ed94500 | 0x44c0000 | +| 28 | blk.2.ffn_norm.weight | 0x53254500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x53259500 | 0x44c0000 | +| 30 | blk.3.attn_k.weight | 0x57719500 | 0x226000 | +| 31 | blk.3.attn_norm.weight | 0x5793f500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x57944500 | 0xb40000 | +| 33 | blk.3.attn_q.weight | 0x58484500 | 0x898000 | +| 34 | blk.3.attn_v.weight | 0x58d1c500 | 0x2a8000 | +| 35 | blk.3.ffn_down.weight | 0x58fc4500 | 0x5a00000 | +| 36 | blk.3.ffn_gate.weight | 0x5e9c4500 | 0x44c0000 | +| 37 | blk.3.ffn_norm.weight | 0x62e84500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0x62e89500 | 0x44c0000 | +| 39 | blk.4.attn_k.weight | 0x67349500 | 0x226000 | +| 40 | blk.4.attn_norm.weight | 0x6756f500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0x67574500 | 0xb40000 | +| 42 | blk.4.attn_q.weight | 0x680b4500 | 0x898000 | +| 43 | blk.4.attn_v.weight | 0x6894c500 | 0x2a8000 | +| 44 | blk.4.ffn_down.weight | 0x68bf4500 | 0x5a00000 | +| 45 | blk.4.ffn_gate.weight | 0x6e5f4500 | 0x44c0000 | +| 46 | blk.4.ffn_norm.weight | 0x72ab4500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0x72ab9500 | 0x44c0000 | +| 48 | blk.5.attn_k.weight | 0x76f79500 | 0x226000 | +| 49 | blk.5.attn_norm.weight | 0x7719f500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0x771a4500 | 0xb40000 | +| 51 | blk.5.attn_q.weight | 0x77ce4500 | 0x898000 | +| 52 | blk.5.attn_v.weight | 0x7857c500 | 0x2a8000 | +| 53 | blk.5.ffn_down.weight | 0x78824500 | 0x5a00000 | +| 54 | blk.5.ffn_gate.weight | 0x7e224500 | 0x44c0000 | +| 55 | blk.5.ffn_norm.weight | 0x826e4500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0x826e9500 | 0x44c0000 | +| 57 | blk.6.attn_k.weight | 0x86ba9500 | 0x226000 | +| 58 | blk.6.attn_norm.weight | 0x86dcf500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0x86dd4500 | 0xb40000 | +| 60 | blk.6.attn_q.weight | 0x87914500 | 0x898000 | +| 61 | blk.6.attn_v.weight | 0x881ac500 | 0x2a8000 | +| 62 | blk.6.ffn_down.weight | 0x88454500 | 0x5a00000 | +| 63 | blk.6.ffn_gate.weight | 0x8de54500 | 0x44c0000 | +| 64 | blk.6.ffn_norm.weight | 0x92314500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0x92319500 | 0x44c0000 | +| 66 | blk.7.attn_k.weight | 0x967d9500 | 0x226000 | +| 67 | blk.7.attn_norm.weight | 0x969ff500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0x96a04500 | 0xb40000 | +| 69 | blk.7.attn_q.weight | 0x97544500 | 0x898000 | +| 70 | blk.7.attn_v.weight | 0x97ddc500 | 0x2a8000 | +| 71 | blk.7.ffn_down.weight | 0x98084500 | 0x5a00000 | +| 72 | blk.7.ffn_gate.weight | 0x9da84500 | 0x44c0000 | +| 73 | blk.7.ffn_norm.weight | 0xa1f44500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0xa1f49500 | 0x44c0000 | +| 75 | blk.8.attn_k.weight | 0xa6409500 | 0x226000 | +| 76 | blk.8.attn_norm.weight | 0xa662f500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0xa6634500 | 0xb40000 | +| 78 | blk.8.attn_q.weight | 0xa7174500 | 0x898000 | +| 79 | blk.8.attn_v.weight | 0xa7a0c500 | 0x2a8000 | +| 80 | blk.8.ffn_down.weight | 0xa7cb4500 | 0x5a00000 | +| 81 | blk.8.ffn_gate.weight | 0xad6b4500 | 0x44c0000 | +| 82 | blk.8.ffn_norm.weight | 0xb1b74500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0xb1b79500 | 0x44c0000 | +| 84 | blk.9.attn_k.weight | 0xb6039500 | 0x226000 | +| 85 | blk.9.attn_norm.weight | 0xb625f500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0xb6264500 | 0xb40000 | +| 87 | blk.9.attn_q.weight | 0xb6da4500 | 0x898000 | +| 88 | blk.9.attn_v.weight | 0xb763c500 | 0x2a8000 | +| 89 | blk.9.ffn_down.weight | 0xb78e4500 | 0x5a00000 | +| 90 | blk.9.ffn_gate.weight | 0xbd2e4500 | 0x44c0000 | +| 91 | blk.9.ffn_norm.weight | 0xc17a4500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0xc17a9500 | 0x44c0000 | +| 93 | blk.10.attn_k.weight | 0xc5c69500 | 0x226000 | +| 94 | blk.10.attn_norm.weight | 0xc5e8f500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0xc5e94500 | 0xb40000 | +| 96 | blk.10.attn_q.weight | 0xc69d4500 | 0x898000 | +| 97 | blk.10.attn_v.weight | 0xc726c500 | 0x2a8000 | +| 98 | blk.10.ffn_down.weight | 0xc7514500 | 0x5a00000 | +| 99 | blk.10.ffn_gate.weight | 0xccf14500 | 0x44c0000 | +| 100 | blk.10.ffn_norm.weight | 0xd13d4500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0xd13d9500 | 0x44c0000 | +| 102 | blk.11.attn_k.weight | 0xd5899500 | 0x226000 | +| 103 | blk.11.attn_norm.weight | 0xd5abf500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0xd5ac4500 | 0xb40000 | +| 105 | blk.11.attn_q.weight | 0xd6604500 | 0x898000 | +| 106 | blk.11.attn_v.weight | 0xd6e9c500 | 0x2a8000 | +| 107 | blk.11.ffn_down.weight | 0xd7144500 | 0x5a00000 | +| 108 | blk.11.ffn_gate.weight | 0xdcb44500 | 0x44c0000 | +| 109 | blk.11.ffn_norm.weight | 0xe1004500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0xe1009500 | 0x44c0000 | +| 111 | blk.12.attn_k.weight | 0xe54c9500 | 0x226000 | +| 112 | blk.12.attn_norm.weight | 0xe56ef500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0xe56f4500 | 0xb40000 | +| 114 | blk.12.attn_q.weight | 0xe6234500 | 0x898000 | +| 115 | blk.12.attn_v.weight | 0xe6acc500 | 0x2a8000 | +| 116 | blk.12.ffn_down.weight | 0xe6d74500 | 0x5a00000 | +| 117 | blk.12.ffn_gate.weight | 0xec774500 | 0x44c0000 | +| 118 | blk.12.ffn_norm.weight | 0xf0c34500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0xf0c39500 | 0x44c0000 | +| 120 | blk.13.attn_k.weight | 0xf50f9500 | 0x226000 | +| 121 | blk.13.attn_norm.weight | 0xf531f500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0xf5324500 | 0xb40000 | +| 123 | blk.13.attn_q.weight | 0xf5e64500 | 0x898000 | +| 124 | blk.13.attn_v.weight | 0xf66fc500 | 0x2a8000 | +| 125 | blk.13.ffn_down.weight | 0xf69a4500 | 0x5a00000 | +| 126 | blk.13.ffn_gate.weight | 0xfc3a4500 | 0x44c0000 | +| 127 | blk.13.ffn_norm.weight | 0x100864500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0x100869500 | 0x44c0000 | +| 129 | blk.14.attn_k.weight | 0x104d29500 | 0x226000 | +| 130 | blk.14.attn_norm.weight | 0x104f4f500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0x104f54500 | 0xb40000 | +| 132 | blk.14.attn_q.weight | 0x105a94500 | 0x898000 | +| 133 | blk.14.attn_v.weight | 0x10632c500 | 0x2a8000 | +| 134 | blk.14.ffn_down.weight | 0x1065d4500 | 0x5a00000 | +| 135 | blk.14.ffn_gate.weight | 0x10bfd4500 | 0x44c0000 | +| 136 | blk.14.ffn_norm.weight | 0x110494500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0x110499500 | 0x44c0000 | +| 138 | blk.15.attn_k.weight | 0x114959500 | 0x226000 | +| 139 | blk.15.attn_norm.weight | 0x114b7f500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0x114b84500 | 0xb40000 | +| 141 | blk.15.attn_q.weight | 0x1156c4500 | 0x898000 | +| 142 | blk.15.attn_v.weight | 0x115f5c500 | 0x2a8000 | +| 143 | blk.15.ffn_down.weight | 0x116204500 | 0x5a00000 | +| 144 | blk.15.ffn_gate.weight | 0x11bc04500 | 0x44c0000 | +| 145 | blk.15.ffn_norm.weight | 0x1200c4500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0x1200c9500 | 0x44c0000 | +| 147 | blk.16.attn_k.weight | 0x124589500 | 0x226000 | +| 148 | blk.16.attn_norm.weight | 0x1247af500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0x1247b4500 | 0xb40000 | +| 150 | blk.16.attn_q.weight | 0x1252f4500 | 0x898000 | +| 151 | blk.16.attn_v.weight | 0x125b8c500 | 0x2a8000 | +| 152 | blk.16.ffn_down.weight | 0x125e34500 | 0x5a00000 | +| 153 | blk.16.ffn_gate.weight | 0x12b834500 | 0x44c0000 | +| 154 | blk.16.ffn_norm.weight | 0x12fcf4500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0x12fcf9500 | 0x44c0000 | +| 156 | blk.17.attn_k.weight | 0x1341b9500 | 0x2d0000 | +| 157 | blk.17.attn_norm.weight | 0x134489500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0x13448e500 | 0xb40000 | +| 159 | blk.17.attn_q.weight | 0x134fce500 | 0xb40000 | +| 160 | blk.17.attn_v.weight | 0x135b0e500 | 0x2d0000 | +| 161 | blk.17.ffn_down.weight | 0x135dde500 | 0x5a00000 | +| 162 | blk.17.ffn_gate.weight | 0x13b7de500 | 0x44c0000 | +| 163 | blk.17.ffn_norm.weight | 0x13fc9e500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0x13fca3500 | 0x44c0000 | +| 165 | blk.18.attn_k.weight | 0x144163500 | 0x2d0000 | +| 166 | blk.18.attn_norm.weight | 0x144433500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0x144438500 | 0xb40000 | +| 168 | blk.18.attn_q.weight | 0x144f78500 | 0xb40000 | +| 169 | blk.18.attn_v.weight | 0x145ab8500 | 0x2d0000 | +| 170 | blk.18.ffn_down.weight | 0x145d88500 | 0x5a00000 | +| 171 | blk.18.ffn_gate.weight | 0x14b788500 | 0x44c0000 | +| 172 | blk.18.ffn_norm.weight | 0x14fc48500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x14fc4d500 | 0x44c0000 | +| 174 | blk.19.attn_k.weight | 0x15410d500 | 0x226000 | +| 175 | blk.19.attn_norm.weight | 0x154333500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x154338500 | 0xb40000 | +| 177 | blk.19.attn_q.weight | 0x154e78500 | 0x898000 | +| 178 | blk.19.attn_v.weight | 0x155710500 | 0x2a8000 | +| 179 | blk.19.ffn_down.weight | 0x1559b8500 | 0x5a00000 | +| 180 | blk.19.ffn_gate.weight | 0x15b3b8500 | 0x44c0000 | +| 181 | blk.19.ffn_norm.weight | 0x15f878500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x15f87d500 | 0x44c0000 | +| 183 | blk.20.attn_k.weight | 0x163d3d500 | 0x2d0000 | +| 184 | blk.20.attn_norm.weight | 0x16400d500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x164012500 | 0xb40000 | +| 186 | blk.20.attn_q.weight | 0x164b52500 | 0xb40000 | +| 187 | blk.20.attn_v.weight | 0x165692500 | 0x2d0000 | +| 188 | blk.20.ffn_down.weight | 0x165962500 | 0x5a00000 | +| 189 | blk.20.ffn_gate.weight | 0x16b362500 | 0x5a00000 | +| 190 | blk.20.ffn_norm.weight | 0x170d62500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x170d67500 | 0x5a00000 | +| 192 | blk.21.attn_k.weight | 0x176767500 | 0x226000 | +| 193 | blk.21.attn_norm.weight | 0x17698d500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x176992500 | 0xb40000 | +| 195 | blk.21.attn_q.weight | 0x1774d2500 | 0x898000 | +| 196 | blk.21.attn_v.weight | 0x177d6a500 | 0x2a8000 | +| 197 | blk.21.ffn_down.weight | 0x178012500 | 0x5a00000 | +| 198 | blk.21.ffn_gate.weight | 0x17da12500 | 0x5a00000 | +| 199 | blk.21.ffn_norm.weight | 0x183412500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x183417500 | 0x5a00000 | +| 201 | blk.22.attn_k.weight | 0x188e17500 | 0x2d0000 | +| 202 | blk.22.attn_norm.weight | 0x1890e7500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x1890ec500 | 0xb40000 | +| 204 | blk.22.attn_q.weight | 0x189c2c500 | 0xb40000 | +| 205 | blk.22.attn_v.weight | 0x18a76c500 | 0x2d0000 | +| 206 | blk.22.ffn_down.weight | 0x18aa3c500 | 0x5a00000 | +| 207 | blk.22.ffn_gate.weight | 0x19043c500 | 0x5a00000 | +| 208 | blk.22.ffn_norm.weight | 0x195e3c500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x195e41500 | 0x5a00000 | +| 210 | blk.23.attn_k.weight | 0x19b841500 | 0x2d0000 | +| 211 | blk.23.attn_norm.weight | 0x19bb11500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x19bb16500 | 0xb40000 | +| 213 | blk.23.attn_q.weight | 0x19c656500 | 0xb40000 | +| 214 | blk.23.attn_v.weight | 0x19d196500 | 0x2d0000 | +| 215 | blk.23.ffn_down.weight | 0x19d466500 | 0x5a00000 | +| 216 | blk.23.ffn_gate.weight | 0x1a2e66500 | 0x5a00000 | +| 217 | blk.23.ffn_norm.weight | 0x1a8866500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x1a886b500 | 0x5a00000 | +| 219 | blk.24.attn_k.weight | 0x1ae26b500 | 0x2d0000 | +| 220 | blk.24.attn_norm.weight | 0x1ae53b500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x1ae540500 | 0xb40000 | +| 222 | blk.24.attn_q.weight | 0x1af080500 | 0xb40000 | +| 223 | blk.24.attn_v.weight | 0x1afbc0500 | 0x2d0000 | +| 224 | blk.24.ffn_down.weight | 0x1afe90500 | 0x5a00000 | +| 225 | blk.24.ffn_gate.weight | 0x1b5890500 | 0x5a00000 | +| 226 | blk.24.ffn_norm.weight | 0x1bb290500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x1bb295500 | 0x5a00000 | +| 228 | blk.25.attn_k.weight | 0x1c0c95500 | 0x2d0000 | +| 229 | blk.25.attn_norm.weight | 0x1c0f65500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x1c0f6a500 | 0xb40000 | +| 231 | blk.25.attn_q.weight | 0x1c1aaa500 | 0xb40000 | +| 232 | blk.25.attn_v.weight | 0x1c25ea500 | 0x2d0000 | +| 233 | blk.25.ffn_down.weight | 0x1c28ba500 | 0x5a00000 | +| 234 | blk.25.ffn_gate.weight | 0x1c82ba500 | 0x5a00000 | +| 235 | blk.25.ffn_norm.weight | 0x1cdcba500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x1cdcbf500 | 0x5a00000 | +| 237 | blk.26.attn_k.weight | 0x1d36bf500 | 0x2d0000 | +| 238 | blk.26.attn_norm.weight | 0x1d398f500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x1d3994500 | 0xb40000 | +| 240 | blk.26.attn_q.weight | 0x1d44d4500 | 0xb40000 | +| 241 | blk.26.attn_v.weight | 0x1d5014500 | 0x2d0000 | +| 242 | blk.26.ffn_down.weight | 0x1d52e4500 | 0x5a00000 | +| 243 | blk.26.ffn_gate.weight | 0x1dace4500 | 0x5a00000 | +| 244 | blk.26.ffn_norm.weight | 0x1e06e4500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x1e06e9500 | 0x5a00000 | +| 246 | blk.27.attn_k.weight | 0x1e60e9500 | 0x226000 | +| 247 | blk.27.attn_norm.weight | 0x1e630f500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x1e6314500 | 0xb40000 | +| 249 | blk.27.attn_q.weight | 0x1e6e54500 | 0x898000 | +| 250 | blk.27.attn_v.weight | 0x1e76ec500 | 0x2a8000 | +| 251 | blk.27.ffn_down.weight | 0x1e7994500 | 0x5a00000 | +| 252 | blk.27.ffn_gate.weight | 0x1ed394500 | 0x5a00000 | +| 253 | blk.27.ffn_norm.weight | 0x1f2d94500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x1f2d99500 | 0x5a00000 | +| 255 | blk.28.attn_k.weight | 0x1f8799500 | 0x2d0000 | +| 256 | blk.28.attn_norm.weight | 0x1f8a69500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x1f8a6e500 | 0xb40000 | +| 258 | blk.28.attn_q.weight | 0x1f95ae500 | 0xb40000 | +| 259 | blk.28.attn_v.weight | 0x1fa0ee500 | 0x2d0000 | +| 260 | blk.28.ffn_down.weight | 0x1fa3be500 | 0x5a00000 | +| 261 | blk.28.ffn_gate.weight | 0x1ffdbe500 | 0x5a00000 | +| 262 | blk.28.ffn_norm.weight | 0x2057be500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x2057c3500 | 0x5a00000 | +| 264 | blk.29.attn_k.weight | 0x20b1c3500 | 0x2d0000 | +| 265 | blk.29.attn_norm.weight | 0x20b493500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x20b498500 | 0xb40000 | +| 267 | blk.29.attn_q.weight | 0x20bfd8500 | 0xb40000 | +| 268 | blk.29.attn_v.weight | 0x20cb18500 | 0x2d0000 | +| 269 | blk.29.ffn_down.weight | 0x20cde8500 | 0x5a00000 | +| 270 | blk.29.ffn_gate.weight | 0x2127e8500 | 0x5a00000 | +| 271 | blk.29.ffn_norm.weight | 0x2181e8500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x2181ed500 | 0x5a00000 | +| 273 | blk.30.attn_k.weight | 0x21dbed500 | 0x2d0000 | +| 274 | blk.30.attn_norm.weight | 0x21debd500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x21dec2500 | 0xb40000 | +| 276 | blk.30.attn_q.weight | 0x21ea02500 | 0xb40000 | +| 277 | blk.30.attn_v.weight | 0x21f542500 | 0x2d0000 | +| 278 | blk.30.ffn_down.weight | 0x21f812500 | 0x5a00000 | +| 279 | blk.30.ffn_gate.weight | 0x225212500 | 0x5a00000 | +| 280 | blk.30.ffn_norm.weight | 0x22ac12500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x22ac17500 | 0x5a00000 | +| 282 | blk.31.attn_k.weight | 0x230617500 | 0x2d0000 | +| 283 | blk.31.attn_norm.weight | 0x2308e7500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x2308ec500 | 0xb40000 | +| 285 | blk.31.attn_q.weight | 0x23142c500 | 0xb40000 | +| 286 | blk.31.attn_v.weight | 0x231f6c500 | 0x2d0000 | +| 287 | blk.31.ffn_down.weight | 0x23223c500 | 0x5a00000 | +| 288 | blk.31.ffn_gate.weight | 0x237c3c500 | 0x5a00000 | +| 289 | blk.31.ffn_norm.weight | 0x23d63c500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x23d641500 | 0x5a00000 | +| 291 | blk.32.attn_k.weight | 0x243041500 | 0x2d0000 | +| 292 | blk.32.attn_norm.weight | 0x243311500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x243316500 | 0xb40000 | +| 294 | blk.32.attn_q.weight | 0x243e56500 | 0xb40000 | +| 295 | blk.32.attn_v.weight | 0x244996500 | 0x2d0000 | +| 296 | blk.32.ffn_down.weight | 0x244c66500 | 0x5a00000 | +| 297 | blk.32.ffn_gate.weight | 0x24a666500 | 0x5a00000 | +| 298 | blk.32.ffn_norm.weight | 0x250066500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x25006b500 | 0x5a00000 | +| 300 | blk.33.attn_k.weight | 0x255a6b500 | 0x2d0000 | +| 301 | blk.33.attn_norm.weight | 0x255d3b500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x255d40500 | 0xb40000 | +| 303 | blk.33.attn_q.weight | 0x256880500 | 0xb40000 | +| 304 | blk.33.attn_v.weight | 0x2573c0500 | 0x2d0000 | +| 305 | blk.33.ffn_down.weight | 0x257690500 | 0x5a00000 | +| 306 | blk.33.ffn_gate.weight | 0x25d090500 | 0x5a00000 | +| 307 | blk.33.ffn_norm.weight | 0x262a90500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x262a95500 | 0x5a00000 | +| 309 | blk.34.attn_k.weight | 0x268495500 | 0x2d0000 | +| 310 | blk.34.attn_norm.weight | 0x268765500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x26876a500 | 0xb40000 | +| 312 | blk.34.attn_q.weight | 0x2692aa500 | 0xb40000 | +| 313 | blk.34.attn_v.weight | 0x269dea500 | 0x2d0000 | +| 314 | blk.34.ffn_down.weight | 0x26a0ba500 | 0x5a00000 | +| 315 | blk.34.ffn_gate.weight | 0x26faba500 | 0x5a00000 | +| 316 | blk.34.ffn_norm.weight | 0x2754ba500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x2754bf500 | 0x5a00000 | +| 318 | blk.35.attn_k.weight | 0x27aebf500 | 0x2d0000 | +| 319 | blk.35.attn_norm.weight | 0x27b18f500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x27b194500 | 0xb40000 | +| 321 | blk.35.attn_q.weight | 0x27bcd4500 | 0xb40000 | +| 322 | blk.35.attn_v.weight | 0x27c814500 | 0x2d0000 | +| 323 | blk.35.ffn_down.weight | 0x27cae4500 | 0x5a00000 | +| 324 | blk.35.ffn_gate.weight | 0x2824e4500 | 0x5a00000 | +| 325 | blk.35.ffn_norm.weight | 0x287ee4500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x287ee9500 | 0x5a00000 | +| 327 | blk.36.attn_k.weight | 0x28d8e9500 | 0x2d0000 | +| 328 | blk.36.attn_norm.weight | 0x28dbb9500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x28dbbe500 | 0xb40000 | +| 330 | blk.36.attn_q.weight | 0x28e6fe500 | 0xb40000 | +| 331 | blk.36.attn_v.weight | 0x28f23e500 | 0x2d0000 | +| 332 | blk.36.ffn_down.weight | 0x28f50e500 | 0x5a00000 | +| 333 | blk.36.ffn_gate.weight | 0x294f0e500 | 0x5a00000 | +| 334 | blk.36.ffn_norm.weight | 0x29a90e500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x29a913500 | 0x5a00000 | +| 336 | blk.37.attn_k.weight | 0x2a0313500 | 0x2d0000 | +| 337 | blk.37.attn_norm.weight | 0x2a05e3500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x2a05e8500 | 0xb40000 | +| 339 | blk.37.attn_q.weight | 0x2a1128500 | 0xb40000 | +| 340 | blk.37.attn_v.weight | 0x2a1c68500 | 0x2d0000 | +| 341 | blk.37.ffn_down.weight | 0x2a1f38500 | 0x5a00000 | +| 342 | blk.37.ffn_gate.weight | 0x2a7938500 | 0x5a00000 | +| 343 | blk.37.ffn_norm.weight | 0x2ad338500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x2ad33d500 | 0x5a00000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-------| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | IQ4_NL | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | IQ3_S | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-------| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-------| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-------| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-------| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-------| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-------| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-------| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-------| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-------| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-------| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ3_S | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ3_S | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ3_S | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_XS | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-------| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | IQ4_NL | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | IQ4_NL | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | IQ4_NL | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | IQ4_NL | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | IQ4_NL | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-Q3_K_L.md b/scores/Dolphin-Mistral-24B-Venice-Edition-Q3_K_L.md new file mode 100644 index 0000000..7912dfb --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-Q3_K_L.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-Q3_K_L.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 13 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-Q3\_K\_L.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-q3_k_lgguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0x11300000 | +| 1 | output_norm.weight | 0x11a84500 | 0x5000 | +| 2 | token_embd.weight | 0x11a89500 | 0x11300000 | +| 3 | blk.0.attn_k.weight | 0x22d89500 | 0x1a4000 | +| 4 | blk.0.attn_norm.weight | 0x22f2d500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x22f32500 | 0xdc0000 | +| 6 | blk.0.attn_q.weight | 0x23cf2500 | 0x690000 | +| 7 | blk.0.attn_v.weight | 0x24382500 | 0x2d0000 | +| 8 | blk.0.ffn_down.weight | 0x24652500 | 0x6e00000 | +| 9 | blk.0.ffn_gate.weight | 0x2b452500 | 0x3480000 | +| 10 | blk.0.ffn_norm.weight | 0x2e8d2500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x2e8d7500 | 0x3480000 | +| 12 | blk.1.attn_k.weight | 0x31d57500 | 0x1a4000 | +| 13 | blk.1.attn_norm.weight | 0x31efb500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x31f00500 | 0xdc0000 | +| 15 | blk.1.attn_q.weight | 0x32cc0500 | 0x690000 | +| 16 | blk.1.attn_v.weight | 0x33350500 | 0x2d0000 | +| 17 | blk.1.ffn_down.weight | 0x33620500 | 0x6e00000 | +| 18 | blk.1.ffn_gate.weight | 0x3a420500 | 0x3480000 | +| 19 | blk.1.ffn_norm.weight | 0x3d8a0500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x3d8a5500 | 0x3480000 | +| 21 | blk.2.attn_k.weight | 0x40d25500 | 0x1a4000 | +| 22 | blk.2.attn_norm.weight | 0x40ec9500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x40ece500 | 0xdc0000 | +| 24 | blk.2.attn_q.weight | 0x41c8e500 | 0x690000 | +| 25 | blk.2.attn_v.weight | 0x4231e500 | 0x2d0000 | +| 26 | blk.2.ffn_down.weight | 0x425ee500 | 0x6e00000 | +| 27 | blk.2.ffn_gate.weight | 0x493ee500 | 0x3480000 | +| 28 | blk.2.ffn_norm.weight | 0x4c86e500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x4c873500 | 0x3480000 | +| 30 | blk.3.attn_k.weight | 0x4fcf3500 | 0x1a4000 | +| 31 | blk.3.attn_norm.weight | 0x4fe97500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x4fe9c500 | 0xdc0000 | +| 33 | blk.3.attn_q.weight | 0x50c5c500 | 0x690000 | +| 34 | blk.3.attn_v.weight | 0x512ec500 | 0x2d0000 | +| 35 | blk.3.ffn_down.weight | 0x515bc500 | 0x6e00000 | +| 36 | blk.3.ffn_gate.weight | 0x583bc500 | 0x3480000 | +| 37 | blk.3.ffn_norm.weight | 0x5b83c500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0x5b841500 | 0x3480000 | +| 39 | blk.4.attn_k.weight | 0x5ecc1500 | 0x1a4000 | +| 40 | blk.4.attn_norm.weight | 0x5ee65500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0x5ee6a500 | 0xdc0000 | +| 42 | blk.4.attn_q.weight | 0x5fc2a500 | 0x690000 | +| 43 | blk.4.attn_v.weight | 0x602ba500 | 0x2d0000 | +| 44 | blk.4.ffn_down.weight | 0x6058a500 | 0x6e00000 | +| 45 | blk.4.ffn_gate.weight | 0x6738a500 | 0x3480000 | +| 46 | blk.4.ffn_norm.weight | 0x6a80a500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0x6a80f500 | 0x3480000 | +| 48 | blk.5.attn_k.weight | 0x6dc8f500 | 0x1a4000 | +| 49 | blk.5.attn_norm.weight | 0x6de33500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0x6de38500 | 0xdc0000 | +| 51 | blk.5.attn_q.weight | 0x6ebf8500 | 0x690000 | +| 52 | blk.5.attn_v.weight | 0x6f288500 | 0x2d0000 | +| 53 | blk.5.ffn_down.weight | 0x6f558500 | 0x6e00000 | +| 54 | blk.5.ffn_gate.weight | 0x76358500 | 0x3480000 | +| 55 | blk.5.ffn_norm.weight | 0x797d8500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0x797dd500 | 0x3480000 | +| 57 | blk.6.attn_k.weight | 0x7cc5d500 | 0x1a4000 | +| 58 | blk.6.attn_norm.weight | 0x7ce01500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0x7ce06500 | 0xdc0000 | +| 60 | blk.6.attn_q.weight | 0x7dbc6500 | 0x690000 | +| 61 | blk.6.attn_v.weight | 0x7e256500 | 0x2d0000 | +| 62 | blk.6.ffn_down.weight | 0x7e526500 | 0x6e00000 | +| 63 | blk.6.ffn_gate.weight | 0x85326500 | 0x3480000 | +| 64 | blk.6.ffn_norm.weight | 0x887a6500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0x887ab500 | 0x3480000 | +| 66 | blk.7.attn_k.weight | 0x8bc2b500 | 0x1a4000 | +| 67 | blk.7.attn_norm.weight | 0x8bdcf500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0x8bdd4500 | 0xdc0000 | +| 69 | blk.7.attn_q.weight | 0x8cb94500 | 0x690000 | +| 70 | blk.7.attn_v.weight | 0x8d224500 | 0x2d0000 | +| 71 | blk.7.ffn_down.weight | 0x8d4f4500 | 0x6e00000 | +| 72 | blk.7.ffn_gate.weight | 0x942f4500 | 0x3480000 | +| 73 | blk.7.ffn_norm.weight | 0x97774500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0x97779500 | 0x3480000 | +| 75 | blk.8.attn_k.weight | 0x9abf9500 | 0x1a4000 | +| 76 | blk.8.attn_norm.weight | 0x9ad9d500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0x9ada2500 | 0xdc0000 | +| 78 | blk.8.attn_q.weight | 0x9bb62500 | 0x690000 | +| 79 | blk.8.attn_v.weight | 0x9c1f2500 | 0x2d0000 | +| 80 | blk.8.ffn_down.weight | 0x9c4c2500 | 0x6e00000 | +| 81 | blk.8.ffn_gate.weight | 0xa32c2500 | 0x3480000 | +| 82 | blk.8.ffn_norm.weight | 0xa6742500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0xa6747500 | 0x3480000 | +| 84 | blk.9.attn_k.weight | 0xa9bc7500 | 0x1a4000 | +| 85 | blk.9.attn_norm.weight | 0xa9d6b500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0xa9d70500 | 0xdc0000 | +| 87 | blk.9.attn_q.weight | 0xaab30500 | 0x690000 | +| 88 | blk.9.attn_v.weight | 0xab1c0500 | 0x2d0000 | +| 89 | blk.9.ffn_down.weight | 0xab490500 | 0x6e00000 | +| 90 | blk.9.ffn_gate.weight | 0xb2290500 | 0x3480000 | +| 91 | blk.9.ffn_norm.weight | 0xb5710500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0xb5715500 | 0x3480000 | +| 93 | blk.10.attn_k.weight | 0xb8b95500 | 0x1a4000 | +| 94 | blk.10.attn_norm.weight | 0xb8d39500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0xb8d3e500 | 0xdc0000 | +| 96 | blk.10.attn_q.weight | 0xb9afe500 | 0x690000 | +| 97 | blk.10.attn_v.weight | 0xba18e500 | 0x2d0000 | +| 98 | blk.10.ffn_down.weight | 0xba45e500 | 0x6e00000 | +| 99 | blk.10.ffn_gate.weight | 0xc125e500 | 0x3480000 | +| 100 | blk.10.ffn_norm.weight | 0xc46de500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0xc46e3500 | 0x3480000 | +| 102 | blk.11.attn_k.weight | 0xc7b63500 | 0x1a4000 | +| 103 | blk.11.attn_norm.weight | 0xc7d07500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0xc7d0c500 | 0xdc0000 | +| 105 | blk.11.attn_q.weight | 0xc8acc500 | 0x690000 | +| 106 | blk.11.attn_v.weight | 0xc915c500 | 0x2d0000 | +| 107 | blk.11.ffn_down.weight | 0xc942c500 | 0x6e00000 | +| 108 | blk.11.ffn_gate.weight | 0xd022c500 | 0x3480000 | +| 109 | blk.11.ffn_norm.weight | 0xd36ac500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0xd36b1500 | 0x3480000 | +| 111 | blk.12.attn_k.weight | 0xd6b31500 | 0x1a4000 | +| 112 | blk.12.attn_norm.weight | 0xd6cd5500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0xd6cda500 | 0xdc0000 | +| 114 | blk.12.attn_q.weight | 0xd7a9a500 | 0x690000 | +| 115 | blk.12.attn_v.weight | 0xd812a500 | 0x2d0000 | +| 116 | blk.12.ffn_down.weight | 0xd83fa500 | 0x6e00000 | +| 117 | blk.12.ffn_gate.weight | 0xdf1fa500 | 0x3480000 | +| 118 | blk.12.ffn_norm.weight | 0xe267a500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0xe267f500 | 0x3480000 | +| 120 | blk.13.attn_k.weight | 0xe5aff500 | 0x1a4000 | +| 121 | blk.13.attn_norm.weight | 0xe5ca3500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0xe5ca8500 | 0xdc0000 | +| 123 | blk.13.attn_q.weight | 0xe6a68500 | 0x690000 | +| 124 | blk.13.attn_v.weight | 0xe70f8500 | 0x2d0000 | +| 125 | blk.13.ffn_down.weight | 0xe73c8500 | 0x6e00000 | +| 126 | blk.13.ffn_gate.weight | 0xee1c8500 | 0x3480000 | +| 127 | blk.13.ffn_norm.weight | 0xf1648500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0xf164d500 | 0x3480000 | +| 129 | blk.14.attn_k.weight | 0xf4acd500 | 0x1a4000 | +| 130 | blk.14.attn_norm.weight | 0xf4c71500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0xf4c76500 | 0xdc0000 | +| 132 | blk.14.attn_q.weight | 0xf5a36500 | 0x690000 | +| 133 | blk.14.attn_v.weight | 0xf60c6500 | 0x2d0000 | +| 134 | blk.14.ffn_down.weight | 0xf6396500 | 0x6e00000 | +| 135 | blk.14.ffn_gate.weight | 0xfd196500 | 0x3480000 | +| 136 | blk.14.ffn_norm.weight | 0x100616500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0x10061b500 | 0x3480000 | +| 138 | blk.15.attn_k.weight | 0x103a9b500 | 0x1a4000 | +| 139 | blk.15.attn_norm.weight | 0x103c3f500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0x103c44500 | 0xdc0000 | +| 141 | blk.15.attn_q.weight | 0x104a04500 | 0x690000 | +| 142 | blk.15.attn_v.weight | 0x105094500 | 0x2d0000 | +| 143 | blk.15.ffn_down.weight | 0x105364500 | 0x6e00000 | +| 144 | blk.15.ffn_gate.weight | 0x10c164500 | 0x3480000 | +| 145 | blk.15.ffn_norm.weight | 0x10f5e4500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0x10f5e9500 | 0x3480000 | +| 147 | blk.16.attn_k.weight | 0x112a69500 | 0x1a4000 | +| 148 | blk.16.attn_norm.weight | 0x112c0d500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0x112c12500 | 0xdc0000 | +| 150 | blk.16.attn_q.weight | 0x1139d2500 | 0x690000 | +| 151 | blk.16.attn_v.weight | 0x114062500 | 0x2d0000 | +| 152 | blk.16.ffn_down.weight | 0x114332500 | 0x6e00000 | +| 153 | blk.16.ffn_gate.weight | 0x11b132500 | 0x3480000 | +| 154 | blk.16.ffn_norm.weight | 0x11e5b2500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0x11e5b7500 | 0x3480000 | +| 156 | blk.17.attn_k.weight | 0x121a37500 | 0x226000 | +| 157 | blk.17.attn_norm.weight | 0x121c5d500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0x121c62500 | 0xdc0000 | +| 159 | blk.17.attn_q.weight | 0x122a22500 | 0x898000 | +| 160 | blk.17.attn_v.weight | 0x1232ba500 | 0x2d0000 | +| 161 | blk.17.ffn_down.weight | 0x12358a500 | 0x6e00000 | +| 162 | blk.17.ffn_gate.weight | 0x12a38a500 | 0x3480000 | +| 163 | blk.17.ffn_norm.weight | 0x12d80a500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0x12d80f500 | 0x3480000 | +| 165 | blk.18.attn_k.weight | 0x130c8f500 | 0x226000 | +| 166 | blk.18.attn_norm.weight | 0x130eb5500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0x130eba500 | 0xdc0000 | +| 168 | blk.18.attn_q.weight | 0x131c7a500 | 0x898000 | +| 169 | blk.18.attn_v.weight | 0x132512500 | 0x2d0000 | +| 170 | blk.18.ffn_down.weight | 0x1327e2500 | 0x6e00000 | +| 171 | blk.18.ffn_gate.weight | 0x1395e2500 | 0x3480000 | +| 172 | blk.18.ffn_norm.weight | 0x13ca62500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x13ca67500 | 0x3480000 | +| 174 | blk.19.attn_k.weight | 0x13fee7500 | 0x1a4000 | +| 175 | blk.19.attn_norm.weight | 0x14008b500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x140090500 | 0xdc0000 | +| 177 | blk.19.attn_q.weight | 0x140e50500 | 0x690000 | +| 178 | blk.19.attn_v.weight | 0x1414e0500 | 0x2d0000 | +| 179 | blk.19.ffn_down.weight | 0x1417b0500 | 0x6e00000 | +| 180 | blk.19.ffn_gate.weight | 0x1485b0500 | 0x3480000 | +| 181 | blk.19.ffn_norm.weight | 0x14ba30500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x14ba35500 | 0x3480000 | +| 183 | blk.20.attn_k.weight | 0x14eeb5500 | 0x226000 | +| 184 | blk.20.attn_norm.weight | 0x14f0db500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x14f0e0500 | 0xdc0000 | +| 186 | blk.20.attn_q.weight | 0x14fea0500 | 0x898000 | +| 187 | blk.20.attn_v.weight | 0x150738500 | 0x2d0000 | +| 188 | blk.20.ffn_down.weight | 0x150a08500 | 0x6e00000 | +| 189 | blk.20.ffn_gate.weight | 0x157808500 | 0x44c0000 | +| 190 | blk.20.ffn_norm.weight | 0x15bcc8500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x15bccd500 | 0x44c0000 | +| 192 | blk.21.attn_k.weight | 0x16018d500 | 0x1a4000 | +| 193 | blk.21.attn_norm.weight | 0x160331500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x160336500 | 0xdc0000 | +| 195 | blk.21.attn_q.weight | 0x1610f6500 | 0x690000 | +| 196 | blk.21.attn_v.weight | 0x161786500 | 0x2d0000 | +| 197 | blk.21.ffn_down.weight | 0x161a56500 | 0x6e00000 | +| 198 | blk.21.ffn_gate.weight | 0x168856500 | 0x44c0000 | +| 199 | blk.21.ffn_norm.weight | 0x16cd16500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x16cd1b500 | 0x44c0000 | +| 201 | blk.22.attn_k.weight | 0x1711db500 | 0x226000 | +| 202 | blk.22.attn_norm.weight | 0x171401500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x171406500 | 0xdc0000 | +| 204 | blk.22.attn_q.weight | 0x1721c6500 | 0x898000 | +| 205 | blk.22.attn_v.weight | 0x172a5e500 | 0x2d0000 | +| 206 | blk.22.ffn_down.weight | 0x172d2e500 | 0x6e00000 | +| 207 | blk.22.ffn_gate.weight | 0x179b2e500 | 0x44c0000 | +| 208 | blk.22.ffn_norm.weight | 0x17dfee500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x17dff3500 | 0x44c0000 | +| 210 | blk.23.attn_k.weight | 0x1824b3500 | 0x226000 | +| 211 | blk.23.attn_norm.weight | 0x1826d9500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x1826de500 | 0xdc0000 | +| 213 | blk.23.attn_q.weight | 0x18349e500 | 0x898000 | +| 214 | blk.23.attn_v.weight | 0x183d36500 | 0x2d0000 | +| 215 | blk.23.ffn_down.weight | 0x184006500 | 0x6e00000 | +| 216 | blk.23.ffn_gate.weight | 0x18ae06500 | 0x44c0000 | +| 217 | blk.23.ffn_norm.weight | 0x18f2c6500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x18f2cb500 | 0x44c0000 | +| 219 | blk.24.attn_k.weight | 0x19378b500 | 0x226000 | +| 220 | blk.24.attn_norm.weight | 0x1939b1500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x1939b6500 | 0xdc0000 | +| 222 | blk.24.attn_q.weight | 0x194776500 | 0x898000 | +| 223 | blk.24.attn_v.weight | 0x19500e500 | 0x2d0000 | +| 224 | blk.24.ffn_down.weight | 0x1952de500 | 0x6e00000 | +| 225 | blk.24.ffn_gate.weight | 0x19c0de500 | 0x44c0000 | +| 226 | blk.24.ffn_norm.weight | 0x1a059e500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x1a05a3500 | 0x44c0000 | +| 228 | blk.25.attn_k.weight | 0x1a4a63500 | 0x226000 | +| 229 | blk.25.attn_norm.weight | 0x1a4c89500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x1a4c8e500 | 0xdc0000 | +| 231 | blk.25.attn_q.weight | 0x1a5a4e500 | 0x898000 | +| 232 | blk.25.attn_v.weight | 0x1a62e6500 | 0x2d0000 | +| 233 | blk.25.ffn_down.weight | 0x1a65b6500 | 0x6e00000 | +| 234 | blk.25.ffn_gate.weight | 0x1ad3b6500 | 0x44c0000 | +| 235 | blk.25.ffn_norm.weight | 0x1b1876500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x1b187b500 | 0x44c0000 | +| 237 | blk.26.attn_k.weight | 0x1b5d3b500 | 0x226000 | +| 238 | blk.26.attn_norm.weight | 0x1b5f61500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x1b5f66500 | 0xdc0000 | +| 240 | blk.26.attn_q.weight | 0x1b6d26500 | 0x898000 | +| 241 | blk.26.attn_v.weight | 0x1b75be500 | 0x2d0000 | +| 242 | blk.26.ffn_down.weight | 0x1b788e500 | 0x6e00000 | +| 243 | blk.26.ffn_gate.weight | 0x1be68e500 | 0x44c0000 | +| 244 | blk.26.ffn_norm.weight | 0x1c2b4e500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x1c2b53500 | 0x44c0000 | +| 246 | blk.27.attn_k.weight | 0x1c7013500 | 0x1a4000 | +| 247 | blk.27.attn_norm.weight | 0x1c71b7500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x1c71bc500 | 0xdc0000 | +| 249 | blk.27.attn_q.weight | 0x1c7f7c500 | 0x690000 | +| 250 | blk.27.attn_v.weight | 0x1c860c500 | 0x2d0000 | +| 251 | blk.27.ffn_down.weight | 0x1c88dc500 | 0x6e00000 | +| 252 | blk.27.ffn_gate.weight | 0x1cf6dc500 | 0x44c0000 | +| 253 | blk.27.ffn_norm.weight | 0x1d3b9c500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x1d3ba1500 | 0x44c0000 | +| 255 | blk.28.attn_k.weight | 0x1d8061500 | 0x226000 | +| 256 | blk.28.attn_norm.weight | 0x1d8287500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x1d828c500 | 0xdc0000 | +| 258 | blk.28.attn_q.weight | 0x1d904c500 | 0x898000 | +| 259 | blk.28.attn_v.weight | 0x1d98e4500 | 0x2d0000 | +| 260 | blk.28.ffn_down.weight | 0x1d9bb4500 | 0x6e00000 | +| 261 | blk.28.ffn_gate.weight | 0x1e09b4500 | 0x44c0000 | +| 262 | blk.28.ffn_norm.weight | 0x1e4e74500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x1e4e79500 | 0x44c0000 | +| 264 | blk.29.attn_k.weight | 0x1e9339500 | 0x226000 | +| 265 | blk.29.attn_norm.weight | 0x1e955f500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x1e9564500 | 0xdc0000 | +| 267 | blk.29.attn_q.weight | 0x1ea324500 | 0x898000 | +| 268 | blk.29.attn_v.weight | 0x1eabbc500 | 0x2d0000 | +| 269 | blk.29.ffn_down.weight | 0x1eae8c500 | 0x6e00000 | +| 270 | blk.29.ffn_gate.weight | 0x1f1c8c500 | 0x44c0000 | +| 271 | blk.29.ffn_norm.weight | 0x1f614c500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x1f6151500 | 0x44c0000 | +| 273 | blk.30.attn_k.weight | 0x1fa611500 | 0x226000 | +| 274 | blk.30.attn_norm.weight | 0x1fa837500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x1fa83c500 | 0xdc0000 | +| 276 | blk.30.attn_q.weight | 0x1fb5fc500 | 0x898000 | +| 277 | blk.30.attn_v.weight | 0x1fbe94500 | 0x2d0000 | +| 278 | blk.30.ffn_down.weight | 0x1fc164500 | 0x6e00000 | +| 279 | blk.30.ffn_gate.weight | 0x202f64500 | 0x44c0000 | +| 280 | blk.30.ffn_norm.weight | 0x207424500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x207429500 | 0x44c0000 | +| 282 | blk.31.attn_k.weight | 0x20b8e9500 | 0x226000 | +| 283 | blk.31.attn_norm.weight | 0x20bb0f500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x20bb14500 | 0xdc0000 | +| 285 | blk.31.attn_q.weight | 0x20c8d4500 | 0x898000 | +| 286 | blk.31.attn_v.weight | 0x20d16c500 | 0x2d0000 | +| 287 | blk.31.ffn_down.weight | 0x20d43c500 | 0x6e00000 | +| 288 | blk.31.ffn_gate.weight | 0x21423c500 | 0x44c0000 | +| 289 | blk.31.ffn_norm.weight | 0x2186fc500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x218701500 | 0x44c0000 | +| 291 | blk.32.attn_k.weight | 0x21cbc1500 | 0x226000 | +| 292 | blk.32.attn_norm.weight | 0x21cde7500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x21cdec500 | 0xdc0000 | +| 294 | blk.32.attn_q.weight | 0x21dbac500 | 0x898000 | +| 295 | blk.32.attn_v.weight | 0x21e444500 | 0x2d0000 | +| 296 | blk.32.ffn_down.weight | 0x21e714500 | 0x6e00000 | +| 297 | blk.32.ffn_gate.weight | 0x225514500 | 0x44c0000 | +| 298 | blk.32.ffn_norm.weight | 0x2299d4500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x2299d9500 | 0x44c0000 | +| 300 | blk.33.attn_k.weight | 0x22de99500 | 0x226000 | +| 301 | blk.33.attn_norm.weight | 0x22e0bf500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x22e0c4500 | 0xdc0000 | +| 303 | blk.33.attn_q.weight | 0x22ee84500 | 0x898000 | +| 304 | blk.33.attn_v.weight | 0x22f71c500 | 0x2d0000 | +| 305 | blk.33.ffn_down.weight | 0x22f9ec500 | 0x6e00000 | +| 306 | blk.33.ffn_gate.weight | 0x2367ec500 | 0x44c0000 | +| 307 | blk.33.ffn_norm.weight | 0x23acac500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x23acb1500 | 0x44c0000 | +| 309 | blk.34.attn_k.weight | 0x23f171500 | 0x226000 | +| 310 | blk.34.attn_norm.weight | 0x23f397500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x23f39c500 | 0xdc0000 | +| 312 | blk.34.attn_q.weight | 0x24015c500 | 0x898000 | +| 313 | blk.34.attn_v.weight | 0x2409f4500 | 0x2d0000 | +| 314 | blk.34.ffn_down.weight | 0x240cc4500 | 0x6e00000 | +| 315 | blk.34.ffn_gate.weight | 0x247ac4500 | 0x44c0000 | +| 316 | blk.34.ffn_norm.weight | 0x24bf84500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x24bf89500 | 0x44c0000 | +| 318 | blk.35.attn_k.weight | 0x250449500 | 0x226000 | +| 319 | blk.35.attn_norm.weight | 0x25066f500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x250674500 | 0xdc0000 | +| 321 | blk.35.attn_q.weight | 0x251434500 | 0x898000 | +| 322 | blk.35.attn_v.weight | 0x251ccc500 | 0x2d0000 | +| 323 | blk.35.ffn_down.weight | 0x251f9c500 | 0x6e00000 | +| 324 | blk.35.ffn_gate.weight | 0x258d9c500 | 0x44c0000 | +| 325 | blk.35.ffn_norm.weight | 0x25d25c500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x25d261500 | 0x44c0000 | +| 327 | blk.36.attn_k.weight | 0x261721500 | 0x226000 | +| 328 | blk.36.attn_norm.weight | 0x261947500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x26194c500 | 0xdc0000 | +| 330 | blk.36.attn_q.weight | 0x26270c500 | 0x898000 | +| 331 | blk.36.attn_v.weight | 0x262fa4500 | 0x2d0000 | +| 332 | blk.36.ffn_down.weight | 0x263274500 | 0x6e00000 | +| 333 | blk.36.ffn_gate.weight | 0x26a074500 | 0x44c0000 | +| 334 | blk.36.ffn_norm.weight | 0x26e534500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x26e539500 | 0x44c0000 | +| 336 | blk.37.attn_k.weight | 0x2729f9500 | 0x226000 | +| 337 | blk.37.attn_norm.weight | 0x272c1f500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x272c24500 | 0xdc0000 | +| 339 | blk.37.attn_q.weight | 0x2739e4500 | 0x898000 | +| 340 | blk.37.attn_v.weight | 0x27427c500 | 0x2d0000 | +| 341 | blk.37.ffn_down.weight | 0x27454c500 | 0x6e00000 | +| 342 | blk.37.ffn_gate.weight | 0x27b34c500 | 0x44c0000 | +| 343 | blk.37.ffn_norm.weight | 0x27f80c500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x27f811500 | 0x44c0000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-Q3_K_M.md b/scores/Dolphin-Mistral-24B-Venice-Edition-Q3_K_M.md new file mode 100644 index 0000000..e2b31f3 --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-Q3_K_M.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-Q3_K_M.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 12 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-Q3\_K\_M.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-q3_k_mgguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0x11300000 | +| 1 | output_norm.weight | 0x11a84500 | 0x5000 | +| 2 | token_embd.weight | 0x11a89500 | 0x11300000 | +| 3 | blk.0.attn_k.weight | 0x22d89500 | 0x1a4000 | +| 4 | blk.0.attn_norm.weight | 0x22f2d500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x22f32500 | 0xb40000 | +| 6 | blk.0.attn_q.weight | 0x23a72500 | 0x690000 | +| 7 | blk.0.attn_v.weight | 0x24102500 | 0x226000 | +| 8 | blk.0.ffn_down.weight | 0x24328500 | 0x6e00000 | +| 9 | blk.0.ffn_gate.weight | 0x2b128500 | 0x3480000 | +| 10 | blk.0.ffn_norm.weight | 0x2e5a8500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x2e5ad500 | 0x3480000 | +| 12 | blk.1.attn_k.weight | 0x31a2d500 | 0x1a4000 | +| 13 | blk.1.attn_norm.weight | 0x31bd1500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x31bd6500 | 0xb40000 | +| 15 | blk.1.attn_q.weight | 0x32716500 | 0x690000 | +| 16 | blk.1.attn_v.weight | 0x32da6500 | 0x226000 | +| 17 | blk.1.ffn_down.weight | 0x32fcc500 | 0x6e00000 | +| 18 | blk.1.ffn_gate.weight | 0x39dcc500 | 0x3480000 | +| 19 | blk.1.ffn_norm.weight | 0x3d24c500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x3d251500 | 0x3480000 | +| 21 | blk.2.attn_k.weight | 0x406d1500 | 0x1a4000 | +| 22 | blk.2.attn_norm.weight | 0x40875500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x4087a500 | 0xb40000 | +| 24 | blk.2.attn_q.weight | 0x413ba500 | 0x690000 | +| 25 | blk.2.attn_v.weight | 0x41a4a500 | 0x226000 | +| 26 | blk.2.ffn_down.weight | 0x41c70500 | 0x5a00000 | +| 27 | blk.2.ffn_gate.weight | 0x47670500 | 0x3480000 | +| 28 | blk.2.ffn_norm.weight | 0x4aaf0500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x4aaf5500 | 0x3480000 | +| 30 | blk.3.attn_k.weight | 0x4df75500 | 0x1a4000 | +| 31 | blk.3.attn_norm.weight | 0x4e119500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x4e11e500 | 0xb40000 | +| 33 | blk.3.attn_q.weight | 0x4ec5e500 | 0x690000 | +| 34 | blk.3.attn_v.weight | 0x4f2ee500 | 0x226000 | +| 35 | blk.3.ffn_down.weight | 0x4f514500 | 0x5a00000 | +| 36 | blk.3.ffn_gate.weight | 0x54f14500 | 0x3480000 | +| 37 | blk.3.ffn_norm.weight | 0x58394500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0x58399500 | 0x3480000 | +| 39 | blk.4.attn_k.weight | 0x5b819500 | 0x1a4000 | +| 40 | blk.4.attn_norm.weight | 0x5b9bd500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0x5b9c2500 | 0xb40000 | +| 42 | blk.4.attn_q.weight | 0x5c502500 | 0x690000 | +| 43 | blk.4.attn_v.weight | 0x5cb92500 | 0x226000 | +| 44 | blk.4.ffn_down.weight | 0x5cdb8500 | 0x5a00000 | +| 45 | blk.4.ffn_gate.weight | 0x627b8500 | 0x3480000 | +| 46 | blk.4.ffn_norm.weight | 0x65c38500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0x65c3d500 | 0x3480000 | +| 48 | blk.5.attn_k.weight | 0x690bd500 | 0x1a4000 | +| 49 | blk.5.attn_norm.weight | 0x69261500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0x69266500 | 0xb40000 | +| 51 | blk.5.attn_q.weight | 0x69da6500 | 0x690000 | +| 52 | blk.5.attn_v.weight | 0x6a436500 | 0x226000 | +| 53 | blk.5.ffn_down.weight | 0x6a65c500 | 0x5a00000 | +| 54 | blk.5.ffn_gate.weight | 0x7005c500 | 0x3480000 | +| 55 | blk.5.ffn_norm.weight | 0x734dc500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0x734e1500 | 0x3480000 | +| 57 | blk.6.attn_k.weight | 0x76961500 | 0x1a4000 | +| 58 | blk.6.attn_norm.weight | 0x76b05500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0x76b0a500 | 0xb40000 | +| 60 | blk.6.attn_q.weight | 0x7764a500 | 0x690000 | +| 61 | blk.6.attn_v.weight | 0x77cda500 | 0x226000 | +| 62 | blk.6.ffn_down.weight | 0x77f00500 | 0x5a00000 | +| 63 | blk.6.ffn_gate.weight | 0x7d900500 | 0x3480000 | +| 64 | blk.6.ffn_norm.weight | 0x80d80500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0x80d85500 | 0x3480000 | +| 66 | blk.7.attn_k.weight | 0x84205500 | 0x1a4000 | +| 67 | blk.7.attn_norm.weight | 0x843a9500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0x843ae500 | 0xb40000 | +| 69 | blk.7.attn_q.weight | 0x84eee500 | 0x690000 | +| 70 | blk.7.attn_v.weight | 0x8557e500 | 0x226000 | +| 71 | blk.7.ffn_down.weight | 0x857a4500 | 0x5a00000 | +| 72 | blk.7.ffn_gate.weight | 0x8b1a4500 | 0x3480000 | +| 73 | blk.7.ffn_norm.weight | 0x8e624500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0x8e629500 | 0x3480000 | +| 75 | blk.8.attn_k.weight | 0x91aa9500 | 0x1a4000 | +| 76 | blk.8.attn_norm.weight | 0x91c4d500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0x91c52500 | 0xb40000 | +| 78 | blk.8.attn_q.weight | 0x92792500 | 0x690000 | +| 79 | blk.8.attn_v.weight | 0x92e22500 | 0x226000 | +| 80 | blk.8.ffn_down.weight | 0x93048500 | 0x5a00000 | +| 81 | blk.8.ffn_gate.weight | 0x98a48500 | 0x3480000 | +| 82 | blk.8.ffn_norm.weight | 0x9bec8500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0x9becd500 | 0x3480000 | +| 84 | blk.9.attn_k.weight | 0x9f34d500 | 0x1a4000 | +| 85 | blk.9.attn_norm.weight | 0x9f4f1500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0x9f4f6500 | 0xb40000 | +| 87 | blk.9.attn_q.weight | 0xa0036500 | 0x690000 | +| 88 | blk.9.attn_v.weight | 0xa06c6500 | 0x226000 | +| 89 | blk.9.ffn_down.weight | 0xa08ec500 | 0x5a00000 | +| 90 | blk.9.ffn_gate.weight | 0xa62ec500 | 0x3480000 | +| 91 | blk.9.ffn_norm.weight | 0xa976c500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0xa9771500 | 0x3480000 | +| 93 | blk.10.attn_k.weight | 0xacbf1500 | 0x1a4000 | +| 94 | blk.10.attn_norm.weight | 0xacd95500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0xacd9a500 | 0xb40000 | +| 96 | blk.10.attn_q.weight | 0xad8da500 | 0x690000 | +| 97 | blk.10.attn_v.weight | 0xadf6a500 | 0x226000 | +| 98 | blk.10.ffn_down.weight | 0xae190500 | 0x5a00000 | +| 99 | blk.10.ffn_gate.weight | 0xb3b90500 | 0x3480000 | +| 100 | blk.10.ffn_norm.weight | 0xb7010500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0xb7015500 | 0x3480000 | +| 102 | blk.11.attn_k.weight | 0xba495500 | 0x1a4000 | +| 103 | blk.11.attn_norm.weight | 0xba639500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0xba63e500 | 0xb40000 | +| 105 | blk.11.attn_q.weight | 0xbb17e500 | 0x690000 | +| 106 | blk.11.attn_v.weight | 0xbb80e500 | 0x226000 | +| 107 | blk.11.ffn_down.weight | 0xbba34500 | 0x5a00000 | +| 108 | blk.11.ffn_gate.weight | 0xc1434500 | 0x3480000 | +| 109 | blk.11.ffn_norm.weight | 0xc48b4500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0xc48b9500 | 0x3480000 | +| 111 | blk.12.attn_k.weight | 0xc7d39500 | 0x1a4000 | +| 112 | blk.12.attn_norm.weight | 0xc7edd500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0xc7ee2500 | 0xb40000 | +| 114 | blk.12.attn_q.weight | 0xc8a22500 | 0x690000 | +| 115 | blk.12.attn_v.weight | 0xc90b2500 | 0x226000 | +| 116 | blk.12.ffn_down.weight | 0xc92d8500 | 0x5a00000 | +| 117 | blk.12.ffn_gate.weight | 0xcecd8500 | 0x3480000 | +| 118 | blk.12.ffn_norm.weight | 0xd2158500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0xd215d500 | 0x3480000 | +| 120 | blk.13.attn_k.weight | 0xd55dd500 | 0x1a4000 | +| 121 | blk.13.attn_norm.weight | 0xd5781500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0xd5786500 | 0xb40000 | +| 123 | blk.13.attn_q.weight | 0xd62c6500 | 0x690000 | +| 124 | blk.13.attn_v.weight | 0xd6956500 | 0x226000 | +| 125 | blk.13.ffn_down.weight | 0xd6b7c500 | 0x5a00000 | +| 126 | blk.13.ffn_gate.weight | 0xdc57c500 | 0x3480000 | +| 127 | blk.13.ffn_norm.weight | 0xdf9fc500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0xdfa01500 | 0x3480000 | +| 129 | blk.14.attn_k.weight | 0xe2e81500 | 0x1a4000 | +| 130 | blk.14.attn_norm.weight | 0xe3025500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0xe302a500 | 0xb40000 | +| 132 | blk.14.attn_q.weight | 0xe3b6a500 | 0x690000 | +| 133 | blk.14.attn_v.weight | 0xe41fa500 | 0x226000 | +| 134 | blk.14.ffn_down.weight | 0xe4420500 | 0x5a00000 | +| 135 | blk.14.ffn_gate.weight | 0xe9e20500 | 0x3480000 | +| 136 | blk.14.ffn_norm.weight | 0xed2a0500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0xed2a5500 | 0x3480000 | +| 138 | blk.15.attn_k.weight | 0xf0725500 | 0x1a4000 | +| 139 | blk.15.attn_norm.weight | 0xf08c9500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0xf08ce500 | 0xb40000 | +| 141 | blk.15.attn_q.weight | 0xf140e500 | 0x690000 | +| 142 | blk.15.attn_v.weight | 0xf1a9e500 | 0x226000 | +| 143 | blk.15.ffn_down.weight | 0xf1cc4500 | 0x5a00000 | +| 144 | blk.15.ffn_gate.weight | 0xf76c4500 | 0x3480000 | +| 145 | blk.15.ffn_norm.weight | 0xfab44500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0xfab49500 | 0x3480000 | +| 147 | blk.16.attn_k.weight | 0xfdfc9500 | 0x1a4000 | +| 148 | blk.16.attn_norm.weight | 0xfe16d500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0xfe172500 | 0xb40000 | +| 150 | blk.16.attn_q.weight | 0xfecb2500 | 0x690000 | +| 151 | blk.16.attn_v.weight | 0xff342500 | 0x226000 | +| 152 | blk.16.ffn_down.weight | 0xff568500 | 0x5a00000 | +| 153 | blk.16.ffn_gate.weight | 0x104f68500 | 0x3480000 | +| 154 | blk.16.ffn_norm.weight | 0x1083e8500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0x1083ed500 | 0x3480000 | +| 156 | blk.17.attn_k.weight | 0x10b86d500 | 0x226000 | +| 157 | blk.17.attn_norm.weight | 0x10ba93500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0x10ba98500 | 0xb40000 | +| 159 | blk.17.attn_q.weight | 0x10c5d8500 | 0x898000 | +| 160 | blk.17.attn_v.weight | 0x10ce70500 | 0x2d0000 | +| 161 | blk.17.ffn_down.weight | 0x10d140500 | 0x5a00000 | +| 162 | blk.17.ffn_gate.weight | 0x112b40500 | 0x3480000 | +| 163 | blk.17.ffn_norm.weight | 0x115fc0500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0x115fc5500 | 0x3480000 | +| 165 | blk.18.attn_k.weight | 0x119445500 | 0x226000 | +| 166 | blk.18.attn_norm.weight | 0x11966b500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0x119670500 | 0xb40000 | +| 168 | blk.18.attn_q.weight | 0x11a1b0500 | 0x898000 | +| 169 | blk.18.attn_v.weight | 0x11aa48500 | 0x2d0000 | +| 170 | blk.18.ffn_down.weight | 0x11ad18500 | 0x5a00000 | +| 171 | blk.18.ffn_gate.weight | 0x120718500 | 0x3480000 | +| 172 | blk.18.ffn_norm.weight | 0x123b98500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x123b9d500 | 0x3480000 | +| 174 | blk.19.attn_k.weight | 0x12701d500 | 0x1a4000 | +| 175 | blk.19.attn_norm.weight | 0x1271c1500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x1271c6500 | 0xb40000 | +| 177 | blk.19.attn_q.weight | 0x127d06500 | 0x690000 | +| 178 | blk.19.attn_v.weight | 0x128396500 | 0x226000 | +| 179 | blk.19.ffn_down.weight | 0x1285bc500 | 0x5a00000 | +| 180 | blk.19.ffn_gate.weight | 0x12dfbc500 | 0x3480000 | +| 181 | blk.19.ffn_norm.weight | 0x13143c500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x131441500 | 0x3480000 | +| 183 | blk.20.attn_k.weight | 0x1348c1500 | 0x226000 | +| 184 | blk.20.attn_norm.weight | 0x134ae7500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x134aec500 | 0xb40000 | +| 186 | blk.20.attn_q.weight | 0x13562c500 | 0x898000 | +| 187 | blk.20.attn_v.weight | 0x135ec4500 | 0x2d0000 | +| 188 | blk.20.ffn_down.weight | 0x136194500 | 0x5a00000 | +| 189 | blk.20.ffn_gate.weight | 0x13bb94500 | 0x44c0000 | +| 190 | blk.20.ffn_norm.weight | 0x140054500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x140059500 | 0x44c0000 | +| 192 | blk.21.attn_k.weight | 0x144519500 | 0x1a4000 | +| 193 | blk.21.attn_norm.weight | 0x1446bd500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x1446c2500 | 0xb40000 | +| 195 | blk.21.attn_q.weight | 0x145202500 | 0x690000 | +| 196 | blk.21.attn_v.weight | 0x145892500 | 0x226000 | +| 197 | blk.21.ffn_down.weight | 0x145ab8500 | 0x5a00000 | +| 198 | blk.21.ffn_gate.weight | 0x14b4b8500 | 0x44c0000 | +| 199 | blk.21.ffn_norm.weight | 0x14f978500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x14f97d500 | 0x44c0000 | +| 201 | blk.22.attn_k.weight | 0x153e3d500 | 0x226000 | +| 202 | blk.22.attn_norm.weight | 0x154063500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x154068500 | 0xb40000 | +| 204 | blk.22.attn_q.weight | 0x154ba8500 | 0x898000 | +| 205 | blk.22.attn_v.weight | 0x155440500 | 0x2d0000 | +| 206 | blk.22.ffn_down.weight | 0x155710500 | 0x5a00000 | +| 207 | blk.22.ffn_gate.weight | 0x15b110500 | 0x44c0000 | +| 208 | blk.22.ffn_norm.weight | 0x15f5d0500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x15f5d5500 | 0x44c0000 | +| 210 | blk.23.attn_k.weight | 0x163a95500 | 0x226000 | +| 211 | blk.23.attn_norm.weight | 0x163cbb500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x163cc0500 | 0xb40000 | +| 213 | blk.23.attn_q.weight | 0x164800500 | 0x898000 | +| 214 | blk.23.attn_v.weight | 0x165098500 | 0x2d0000 | +| 215 | blk.23.ffn_down.weight | 0x165368500 | 0x5a00000 | +| 216 | blk.23.ffn_gate.weight | 0x16ad68500 | 0x44c0000 | +| 217 | blk.23.ffn_norm.weight | 0x16f228500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x16f22d500 | 0x44c0000 | +| 219 | blk.24.attn_k.weight | 0x1736ed500 | 0x226000 | +| 220 | blk.24.attn_norm.weight | 0x173913500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x173918500 | 0xb40000 | +| 222 | blk.24.attn_q.weight | 0x174458500 | 0x898000 | +| 223 | blk.24.attn_v.weight | 0x174cf0500 | 0x2d0000 | +| 224 | blk.24.ffn_down.weight | 0x174fc0500 | 0x5a00000 | +| 225 | blk.24.ffn_gate.weight | 0x17a9c0500 | 0x44c0000 | +| 226 | blk.24.ffn_norm.weight | 0x17ee80500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x17ee85500 | 0x44c0000 | +| 228 | blk.25.attn_k.weight | 0x183345500 | 0x226000 | +| 229 | blk.25.attn_norm.weight | 0x18356b500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x183570500 | 0xb40000 | +| 231 | blk.25.attn_q.weight | 0x1840b0500 | 0x898000 | +| 232 | blk.25.attn_v.weight | 0x184948500 | 0x2d0000 | +| 233 | blk.25.ffn_down.weight | 0x184c18500 | 0x5a00000 | +| 234 | blk.25.ffn_gate.weight | 0x18a618500 | 0x44c0000 | +| 235 | blk.25.ffn_norm.weight | 0x18ead8500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x18eadd500 | 0x44c0000 | +| 237 | blk.26.attn_k.weight | 0x192f9d500 | 0x226000 | +| 238 | blk.26.attn_norm.weight | 0x1931c3500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x1931c8500 | 0xb40000 | +| 240 | blk.26.attn_q.weight | 0x193d08500 | 0x898000 | +| 241 | blk.26.attn_v.weight | 0x1945a0500 | 0x2d0000 | +| 242 | blk.26.ffn_down.weight | 0x194870500 | 0x5a00000 | +| 243 | blk.26.ffn_gate.weight | 0x19a270500 | 0x44c0000 | +| 244 | blk.26.ffn_norm.weight | 0x19e730500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x19e735500 | 0x44c0000 | +| 246 | blk.27.attn_k.weight | 0x1a2bf5500 | 0x1a4000 | +| 247 | blk.27.attn_norm.weight | 0x1a2d99500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x1a2d9e500 | 0xb40000 | +| 249 | blk.27.attn_q.weight | 0x1a38de500 | 0x690000 | +| 250 | blk.27.attn_v.weight | 0x1a3f6e500 | 0x226000 | +| 251 | blk.27.ffn_down.weight | 0x1a4194500 | 0x5a00000 | +| 252 | blk.27.ffn_gate.weight | 0x1a9b94500 | 0x44c0000 | +| 253 | blk.27.ffn_norm.weight | 0x1ae054500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x1ae059500 | 0x44c0000 | +| 255 | blk.28.attn_k.weight | 0x1b2519500 | 0x226000 | +| 256 | blk.28.attn_norm.weight | 0x1b273f500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x1b2744500 | 0xb40000 | +| 258 | blk.28.attn_q.weight | 0x1b3284500 | 0x898000 | +| 259 | blk.28.attn_v.weight | 0x1b3b1c500 | 0x2d0000 | +| 260 | blk.28.ffn_down.weight | 0x1b3dec500 | 0x5a00000 | +| 261 | blk.28.ffn_gate.weight | 0x1b97ec500 | 0x44c0000 | +| 262 | blk.28.ffn_norm.weight | 0x1bdcac500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x1bdcb1500 | 0x44c0000 | +| 264 | blk.29.attn_k.weight | 0x1c2171500 | 0x226000 | +| 265 | blk.29.attn_norm.weight | 0x1c2397500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x1c239c500 | 0xb40000 | +| 267 | blk.29.attn_q.weight | 0x1c2edc500 | 0x898000 | +| 268 | blk.29.attn_v.weight | 0x1c3774500 | 0x2d0000 | +| 269 | blk.29.ffn_down.weight | 0x1c3a44500 | 0x5a00000 | +| 270 | blk.29.ffn_gate.weight | 0x1c9444500 | 0x44c0000 | +| 271 | blk.29.ffn_norm.weight | 0x1cd904500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x1cd909500 | 0x44c0000 | +| 273 | blk.30.attn_k.weight | 0x1d1dc9500 | 0x226000 | +| 274 | blk.30.attn_norm.weight | 0x1d1fef500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x1d1ff4500 | 0xb40000 | +| 276 | blk.30.attn_q.weight | 0x1d2b34500 | 0x898000 | +| 277 | blk.30.attn_v.weight | 0x1d33cc500 | 0x2d0000 | +| 278 | blk.30.ffn_down.weight | 0x1d369c500 | 0x5a00000 | +| 279 | blk.30.ffn_gate.weight | 0x1d909c500 | 0x44c0000 | +| 280 | blk.30.ffn_norm.weight | 0x1dd55c500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x1dd561500 | 0x44c0000 | +| 282 | blk.31.attn_k.weight | 0x1e1a21500 | 0x226000 | +| 283 | blk.31.attn_norm.weight | 0x1e1c47500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x1e1c4c500 | 0xb40000 | +| 285 | blk.31.attn_q.weight | 0x1e278c500 | 0x898000 | +| 286 | blk.31.attn_v.weight | 0x1e3024500 | 0x2d0000 | +| 287 | blk.31.ffn_down.weight | 0x1e32f4500 | 0x5a00000 | +| 288 | blk.31.ffn_gate.weight | 0x1e8cf4500 | 0x44c0000 | +| 289 | blk.31.ffn_norm.weight | 0x1ed1b4500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x1ed1b9500 | 0x44c0000 | +| 291 | blk.32.attn_k.weight | 0x1f1679500 | 0x226000 | +| 292 | blk.32.attn_norm.weight | 0x1f189f500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x1f18a4500 | 0xb40000 | +| 294 | blk.32.attn_q.weight | 0x1f23e4500 | 0x898000 | +| 295 | blk.32.attn_v.weight | 0x1f2c7c500 | 0x2d0000 | +| 296 | blk.32.ffn_down.weight | 0x1f2f4c500 | 0x5a00000 | +| 297 | blk.32.ffn_gate.weight | 0x1f894c500 | 0x44c0000 | +| 298 | blk.32.ffn_norm.weight | 0x1fce0c500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x1fce11500 | 0x44c0000 | +| 300 | blk.33.attn_k.weight | 0x2012d1500 | 0x226000 | +| 301 | blk.33.attn_norm.weight | 0x2014f7500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x2014fc500 | 0xb40000 | +| 303 | blk.33.attn_q.weight | 0x20203c500 | 0x898000 | +| 304 | blk.33.attn_v.weight | 0x2028d4500 | 0x2d0000 | +| 305 | blk.33.ffn_down.weight | 0x202ba4500 | 0x5a00000 | +| 306 | blk.33.ffn_gate.weight | 0x2085a4500 | 0x44c0000 | +| 307 | blk.33.ffn_norm.weight | 0x20ca64500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x20ca69500 | 0x44c0000 | +| 309 | blk.34.attn_k.weight | 0x210f29500 | 0x226000 | +| 310 | blk.34.attn_norm.weight | 0x21114f500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x211154500 | 0xb40000 | +| 312 | blk.34.attn_q.weight | 0x211c94500 | 0x898000 | +| 313 | blk.34.attn_v.weight | 0x21252c500 | 0x2d0000 | +| 314 | blk.34.ffn_down.weight | 0x2127fc500 | 0x5a00000 | +| 315 | blk.34.ffn_gate.weight | 0x2181fc500 | 0x44c0000 | +| 316 | blk.34.ffn_norm.weight | 0x21c6bc500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x21c6c1500 | 0x44c0000 | +| 318 | blk.35.attn_k.weight | 0x220b81500 | 0x226000 | +| 319 | blk.35.attn_norm.weight | 0x220da7500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x220dac500 | 0xb40000 | +| 321 | blk.35.attn_q.weight | 0x2218ec500 | 0x898000 | +| 322 | blk.35.attn_v.weight | 0x222184500 | 0x2d0000 | +| 323 | blk.35.ffn_down.weight | 0x222454500 | 0x5a00000 | +| 324 | blk.35.ffn_gate.weight | 0x227e54500 | 0x44c0000 | +| 325 | blk.35.ffn_norm.weight | 0x22c314500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x22c319500 | 0x44c0000 | +| 327 | blk.36.attn_k.weight | 0x2307d9500 | 0x226000 | +| 328 | blk.36.attn_norm.weight | 0x2309ff500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x230a04500 | 0xb40000 | +| 330 | blk.36.attn_q.weight | 0x231544500 | 0x898000 | +| 331 | blk.36.attn_v.weight | 0x231ddc500 | 0x2d0000 | +| 332 | blk.36.ffn_down.weight | 0x2320ac500 | 0x5a00000 | +| 333 | blk.36.ffn_gate.weight | 0x237aac500 | 0x44c0000 | +| 334 | blk.36.ffn_norm.weight | 0x23bf6c500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x23bf71500 | 0x44c0000 | +| 336 | blk.37.attn_k.weight | 0x240431500 | 0x226000 | +| 337 | blk.37.attn_norm.weight | 0x240657500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x24065c500 | 0xb40000 | +| 339 | blk.37.attn_q.weight | 0x24119c500 | 0x898000 | +| 340 | blk.37.attn_v.weight | 0x241a34500 | 0x2d0000 | +| 341 | blk.37.ffn_down.weight | 0x241d04500 | 0x5a00000 | +| 342 | blk.37.ffn_gate.weight | 0x247704500 | 0x44c0000 | +| 343 | blk.37.ffn_norm.weight | 0x24bbc4500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x24bbc9500 | 0x44c0000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-Q3_K_S.md b/scores/Dolphin-Mistral-24B-Venice-Edition-Q3_K_S.md new file mode 100644 index 0000000..a2dffd8 --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-Q3_K_S.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-Q3_K_S.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 11 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-Q3\_K\_S.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-q3_k_sgguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0x11300000 | +| 1 | output_norm.weight | 0x11a84500 | 0x5000 | +| 2 | token_embd.weight | 0x11a89500 | 0xd200000 | +| 3 | blk.0.attn_k.weight | 0x1ec89500 | 0x1a4000 | +| 4 | blk.0.attn_norm.weight | 0x1ee2d500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x1ee32500 | 0x898000 | +| 6 | blk.0.attn_q.weight | 0x1f6ca500 | 0x690000 | +| 7 | blk.0.attn_v.weight | 0x1fd5a500 | 0x226000 | +| 8 | blk.0.ffn_down.weight | 0x1ff80500 | 0x44c0000 | +| 9 | blk.0.ffn_gate.weight | 0x24440500 | 0x3480000 | +| 10 | blk.0.ffn_norm.weight | 0x278c0500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x278c5500 | 0x3480000 | +| 12 | blk.1.attn_k.weight | 0x2ad45500 | 0x1a4000 | +| 13 | blk.1.attn_norm.weight | 0x2aee9500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x2aeee500 | 0x898000 | +| 15 | blk.1.attn_q.weight | 0x2b786500 | 0x690000 | +| 16 | blk.1.attn_v.weight | 0x2be16500 | 0x226000 | +| 17 | blk.1.ffn_down.weight | 0x2c03c500 | 0x44c0000 | +| 18 | blk.1.ffn_gate.weight | 0x304fc500 | 0x3480000 | +| 19 | blk.1.ffn_norm.weight | 0x3397c500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x33981500 | 0x3480000 | +| 21 | blk.2.attn_k.weight | 0x36e01500 | 0x1a4000 | +| 22 | blk.2.attn_norm.weight | 0x36fa5500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x36faa500 | 0x898000 | +| 24 | blk.2.attn_q.weight | 0x37842500 | 0x690000 | +| 25 | blk.2.attn_v.weight | 0x37ed2500 | 0x226000 | +| 26 | blk.2.ffn_down.weight | 0x380f8500 | 0x44c0000 | +| 27 | blk.2.ffn_gate.weight | 0x3c5b8500 | 0x3480000 | +| 28 | blk.2.ffn_norm.weight | 0x3fa38500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x3fa3d500 | 0x3480000 | +| 30 | blk.3.attn_k.weight | 0x42ebd500 | 0x1a4000 | +| 31 | blk.3.attn_norm.weight | 0x43061500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x43066500 | 0x898000 | +| 33 | blk.3.attn_q.weight | 0x438fe500 | 0x690000 | +| 34 | blk.3.attn_v.weight | 0x43f8e500 | 0x226000 | +| 35 | blk.3.ffn_down.weight | 0x441b4500 | 0x44c0000 | +| 36 | blk.3.ffn_gate.weight | 0x48674500 | 0x3480000 | +| 37 | blk.3.ffn_norm.weight | 0x4baf4500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0x4baf9500 | 0x3480000 | +| 39 | blk.4.attn_k.weight | 0x4ef79500 | 0x1a4000 | +| 40 | blk.4.attn_norm.weight | 0x4f11d500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0x4f122500 | 0x898000 | +| 42 | blk.4.attn_q.weight | 0x4f9ba500 | 0x690000 | +| 43 | blk.4.attn_v.weight | 0x5004a500 | 0x226000 | +| 44 | blk.4.ffn_down.weight | 0x50270500 | 0x44c0000 | +| 45 | blk.4.ffn_gate.weight | 0x54730500 | 0x3480000 | +| 46 | blk.4.ffn_norm.weight | 0x57bb0500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0x57bb5500 | 0x3480000 | +| 48 | blk.5.attn_k.weight | 0x5b035500 | 0x1a4000 | +| 49 | blk.5.attn_norm.weight | 0x5b1d9500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0x5b1de500 | 0x898000 | +| 51 | blk.5.attn_q.weight | 0x5ba76500 | 0x690000 | +| 52 | blk.5.attn_v.weight | 0x5c106500 | 0x226000 | +| 53 | blk.5.ffn_down.weight | 0x5c32c500 | 0x44c0000 | +| 54 | blk.5.ffn_gate.weight | 0x607ec500 | 0x3480000 | +| 55 | blk.5.ffn_norm.weight | 0x63c6c500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0x63c71500 | 0x3480000 | +| 57 | blk.6.attn_k.weight | 0x670f1500 | 0x1a4000 | +| 58 | blk.6.attn_norm.weight | 0x67295500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0x6729a500 | 0x898000 | +| 60 | blk.6.attn_q.weight | 0x67b32500 | 0x690000 | +| 61 | blk.6.attn_v.weight | 0x681c2500 | 0x226000 | +| 62 | blk.6.ffn_down.weight | 0x683e8500 | 0x44c0000 | +| 63 | blk.6.ffn_gate.weight | 0x6c8a8500 | 0x3480000 | +| 64 | blk.6.ffn_norm.weight | 0x6fd28500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0x6fd2d500 | 0x3480000 | +| 66 | blk.7.attn_k.weight | 0x731ad500 | 0x1a4000 | +| 67 | blk.7.attn_norm.weight | 0x73351500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0x73356500 | 0x898000 | +| 69 | blk.7.attn_q.weight | 0x73bee500 | 0x690000 | +| 70 | blk.7.attn_v.weight | 0x7427e500 | 0x226000 | +| 71 | blk.7.ffn_down.weight | 0x744a4500 | 0x44c0000 | +| 72 | blk.7.ffn_gate.weight | 0x78964500 | 0x3480000 | +| 73 | blk.7.ffn_norm.weight | 0x7bde4500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0x7bde9500 | 0x3480000 | +| 75 | blk.8.attn_k.weight | 0x7f269500 | 0x1a4000 | +| 76 | blk.8.attn_norm.weight | 0x7f40d500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0x7f412500 | 0x898000 | +| 78 | blk.8.attn_q.weight | 0x7fcaa500 | 0x690000 | +| 79 | blk.8.attn_v.weight | 0x8033a500 | 0x226000 | +| 80 | blk.8.ffn_down.weight | 0x80560500 | 0x44c0000 | +| 81 | blk.8.ffn_gate.weight | 0x84a20500 | 0x3480000 | +| 82 | blk.8.ffn_norm.weight | 0x87ea0500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0x87ea5500 | 0x3480000 | +| 84 | blk.9.attn_k.weight | 0x8b325500 | 0x1a4000 | +| 85 | blk.9.attn_norm.weight | 0x8b4c9500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0x8b4ce500 | 0x898000 | +| 87 | blk.9.attn_q.weight | 0x8bd66500 | 0x690000 | +| 88 | blk.9.attn_v.weight | 0x8c3f6500 | 0x226000 | +| 89 | blk.9.ffn_down.weight | 0x8c61c500 | 0x44c0000 | +| 90 | blk.9.ffn_gate.weight | 0x90adc500 | 0x3480000 | +| 91 | blk.9.ffn_norm.weight | 0x93f5c500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0x93f61500 | 0x3480000 | +| 93 | blk.10.attn_k.weight | 0x973e1500 | 0x1a4000 | +| 94 | blk.10.attn_norm.weight | 0x97585500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0x9758a500 | 0x898000 | +| 96 | blk.10.attn_q.weight | 0x97e22500 | 0x690000 | +| 97 | blk.10.attn_v.weight | 0x984b2500 | 0x226000 | +| 98 | blk.10.ffn_down.weight | 0x986d8500 | 0x44c0000 | +| 99 | blk.10.ffn_gate.weight | 0x9cb98500 | 0x3480000 | +| 100 | blk.10.ffn_norm.weight | 0xa0018500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0xa001d500 | 0x3480000 | +| 102 | blk.11.attn_k.weight | 0xa349d500 | 0x1a4000 | +| 103 | blk.11.attn_norm.weight | 0xa3641500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0xa3646500 | 0x898000 | +| 105 | blk.11.attn_q.weight | 0xa3ede500 | 0x690000 | +| 106 | blk.11.attn_v.weight | 0xa456e500 | 0x226000 | +| 107 | blk.11.ffn_down.weight | 0xa4794500 | 0x44c0000 | +| 108 | blk.11.ffn_gate.weight | 0xa8c54500 | 0x3480000 | +| 109 | blk.11.ffn_norm.weight | 0xac0d4500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0xac0d9500 | 0x3480000 | +| 111 | blk.12.attn_k.weight | 0xaf559500 | 0x1a4000 | +| 112 | blk.12.attn_norm.weight | 0xaf6fd500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0xaf702500 | 0x898000 | +| 114 | blk.12.attn_q.weight | 0xaff9a500 | 0x690000 | +| 115 | blk.12.attn_v.weight | 0xb062a500 | 0x226000 | +| 116 | blk.12.ffn_down.weight | 0xb0850500 | 0x44c0000 | +| 117 | blk.12.ffn_gate.weight | 0xb4d10500 | 0x3480000 | +| 118 | blk.12.ffn_norm.weight | 0xb8190500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0xb8195500 | 0x3480000 | +| 120 | blk.13.attn_k.weight | 0xbb615500 | 0x1a4000 | +| 121 | blk.13.attn_norm.weight | 0xbb7b9500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0xbb7be500 | 0x898000 | +| 123 | blk.13.attn_q.weight | 0xbc056500 | 0x690000 | +| 124 | blk.13.attn_v.weight | 0xbc6e6500 | 0x226000 | +| 125 | blk.13.ffn_down.weight | 0xbc90c500 | 0x44c0000 | +| 126 | blk.13.ffn_gate.weight | 0xc0dcc500 | 0x3480000 | +| 127 | blk.13.ffn_norm.weight | 0xc424c500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0xc4251500 | 0x3480000 | +| 129 | blk.14.attn_k.weight | 0xc76d1500 | 0x1a4000 | +| 130 | blk.14.attn_norm.weight | 0xc7875500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0xc787a500 | 0x898000 | +| 132 | blk.14.attn_q.weight | 0xc8112500 | 0x690000 | +| 133 | blk.14.attn_v.weight | 0xc87a2500 | 0x226000 | +| 134 | blk.14.ffn_down.weight | 0xc89c8500 | 0x44c0000 | +| 135 | blk.14.ffn_gate.weight | 0xcce88500 | 0x3480000 | +| 136 | blk.14.ffn_norm.weight | 0xd0308500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0xd030d500 | 0x3480000 | +| 138 | blk.15.attn_k.weight | 0xd378d500 | 0x1a4000 | +| 139 | blk.15.attn_norm.weight | 0xd3931500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0xd3936500 | 0x898000 | +| 141 | blk.15.attn_q.weight | 0xd41ce500 | 0x690000 | +| 142 | blk.15.attn_v.weight | 0xd485e500 | 0x226000 | +| 143 | blk.15.ffn_down.weight | 0xd4a84500 | 0x44c0000 | +| 144 | blk.15.ffn_gate.weight | 0xd8f44500 | 0x3480000 | +| 145 | blk.15.ffn_norm.weight | 0xdc3c4500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0xdc3c9500 | 0x3480000 | +| 147 | blk.16.attn_k.weight | 0xdf849500 | 0x1a4000 | +| 148 | blk.16.attn_norm.weight | 0xdf9ed500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0xdf9f2500 | 0x898000 | +| 150 | blk.16.attn_q.weight | 0xe028a500 | 0x690000 | +| 151 | blk.16.attn_v.weight | 0xe091a500 | 0x226000 | +| 152 | blk.16.ffn_down.weight | 0xe0b40500 | 0x44c0000 | +| 153 | blk.16.ffn_gate.weight | 0xe5000500 | 0x3480000 | +| 154 | blk.16.ffn_norm.weight | 0xe8480500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0xe8485500 | 0x3480000 | +| 156 | blk.17.attn_k.weight | 0xeb905500 | 0x226000 | +| 157 | blk.17.attn_norm.weight | 0xebb2b500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0xebb30500 | 0x898000 | +| 159 | blk.17.attn_q.weight | 0xec3c8500 | 0x898000 | +| 160 | blk.17.attn_v.weight | 0xecc60500 | 0x226000 | +| 161 | blk.17.ffn_down.weight | 0xece86500 | 0x44c0000 | +| 162 | blk.17.ffn_gate.weight | 0xf1346500 | 0x3480000 | +| 163 | blk.17.ffn_norm.weight | 0xf47c6500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0xf47cb500 | 0x3480000 | +| 165 | blk.18.attn_k.weight | 0xf7c4b500 | 0x226000 | +| 166 | blk.18.attn_norm.weight | 0xf7e71500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0xf7e76500 | 0x898000 | +| 168 | blk.18.attn_q.weight | 0xf870e500 | 0x898000 | +| 169 | blk.18.attn_v.weight | 0xf8fa6500 | 0x226000 | +| 170 | blk.18.ffn_down.weight | 0xf91cc500 | 0x44c0000 | +| 171 | blk.18.ffn_gate.weight | 0xfd68c500 | 0x3480000 | +| 172 | blk.18.ffn_norm.weight | 0x100b0c500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x100b11500 | 0x3480000 | +| 174 | blk.19.attn_k.weight | 0x103f91500 | 0x1a4000 | +| 175 | blk.19.attn_norm.weight | 0x104135500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x10413a500 | 0x898000 | +| 177 | blk.19.attn_q.weight | 0x1049d2500 | 0x690000 | +| 178 | blk.19.attn_v.weight | 0x105062500 | 0x226000 | +| 179 | blk.19.ffn_down.weight | 0x105288500 | 0x44c0000 | +| 180 | blk.19.ffn_gate.weight | 0x109748500 | 0x3480000 | +| 181 | blk.19.ffn_norm.weight | 0x10cbc8500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x10cbcd500 | 0x3480000 | +| 183 | blk.20.attn_k.weight | 0x11004d500 | 0x226000 | +| 184 | blk.20.attn_norm.weight | 0x110273500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x110278500 | 0x898000 | +| 186 | blk.20.attn_q.weight | 0x110b10500 | 0x898000 | +| 187 | blk.20.attn_v.weight | 0x1113a8500 | 0x226000 | +| 188 | blk.20.ffn_down.weight | 0x1115ce500 | 0x44c0000 | +| 189 | blk.20.ffn_gate.weight | 0x115a8e500 | 0x44c0000 | +| 190 | blk.20.ffn_norm.weight | 0x119f4e500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x119f53500 | 0x44c0000 | +| 192 | blk.21.attn_k.weight | 0x11e413500 | 0x1a4000 | +| 193 | blk.21.attn_norm.weight | 0x11e5b7500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x11e5bc500 | 0x898000 | +| 195 | blk.21.attn_q.weight | 0x11ee54500 | 0x690000 | +| 196 | blk.21.attn_v.weight | 0x11f4e4500 | 0x226000 | +| 197 | blk.21.ffn_down.weight | 0x11f70a500 | 0x44c0000 | +| 198 | blk.21.ffn_gate.weight | 0x123bca500 | 0x44c0000 | +| 199 | blk.21.ffn_norm.weight | 0x12808a500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x12808f500 | 0x44c0000 | +| 201 | blk.22.attn_k.weight | 0x12c54f500 | 0x226000 | +| 202 | blk.22.attn_norm.weight | 0x12c775500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x12c77a500 | 0x898000 | +| 204 | blk.22.attn_q.weight | 0x12d012500 | 0x898000 | +| 205 | blk.22.attn_v.weight | 0x12d8aa500 | 0x226000 | +| 206 | blk.22.ffn_down.weight | 0x12dad0500 | 0x44c0000 | +| 207 | blk.22.ffn_gate.weight | 0x131f90500 | 0x44c0000 | +| 208 | blk.22.ffn_norm.weight | 0x136450500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x136455500 | 0x44c0000 | +| 210 | blk.23.attn_k.weight | 0x13a915500 | 0x226000 | +| 211 | blk.23.attn_norm.weight | 0x13ab3b500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x13ab40500 | 0x898000 | +| 213 | blk.23.attn_q.weight | 0x13b3d8500 | 0x898000 | +| 214 | blk.23.attn_v.weight | 0x13bc70500 | 0x226000 | +| 215 | blk.23.ffn_down.weight | 0x13be96500 | 0x44c0000 | +| 216 | blk.23.ffn_gate.weight | 0x140356500 | 0x44c0000 | +| 217 | blk.23.ffn_norm.weight | 0x144816500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x14481b500 | 0x44c0000 | +| 219 | blk.24.attn_k.weight | 0x148cdb500 | 0x226000 | +| 220 | blk.24.attn_norm.weight | 0x148f01500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x148f06500 | 0x898000 | +| 222 | blk.24.attn_q.weight | 0x14979e500 | 0x898000 | +| 223 | blk.24.attn_v.weight | 0x14a036500 | 0x226000 | +| 224 | blk.24.ffn_down.weight | 0x14a25c500 | 0x44c0000 | +| 225 | blk.24.ffn_gate.weight | 0x14e71c500 | 0x44c0000 | +| 226 | blk.24.ffn_norm.weight | 0x152bdc500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x152be1500 | 0x44c0000 | +| 228 | blk.25.attn_k.weight | 0x1570a1500 | 0x226000 | +| 229 | blk.25.attn_norm.weight | 0x1572c7500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x1572cc500 | 0x898000 | +| 231 | blk.25.attn_q.weight | 0x157b64500 | 0x898000 | +| 232 | blk.25.attn_v.weight | 0x1583fc500 | 0x226000 | +| 233 | blk.25.ffn_down.weight | 0x158622500 | 0x44c0000 | +| 234 | blk.25.ffn_gate.weight | 0x15cae2500 | 0x44c0000 | +| 235 | blk.25.ffn_norm.weight | 0x160fa2500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x160fa7500 | 0x44c0000 | +| 237 | blk.26.attn_k.weight | 0x165467500 | 0x226000 | +| 238 | blk.26.attn_norm.weight | 0x16568d500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x165692500 | 0x898000 | +| 240 | blk.26.attn_q.weight | 0x165f2a500 | 0x898000 | +| 241 | blk.26.attn_v.weight | 0x1667c2500 | 0x226000 | +| 242 | blk.26.ffn_down.weight | 0x1669e8500 | 0x44c0000 | +| 243 | blk.26.ffn_gate.weight | 0x16aea8500 | 0x44c0000 | +| 244 | blk.26.ffn_norm.weight | 0x16f368500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x16f36d500 | 0x44c0000 | +| 246 | blk.27.attn_k.weight | 0x17382d500 | 0x1a4000 | +| 247 | blk.27.attn_norm.weight | 0x1739d1500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x1739d6500 | 0x898000 | +| 249 | blk.27.attn_q.weight | 0x17426e500 | 0x690000 | +| 250 | blk.27.attn_v.weight | 0x1748fe500 | 0x226000 | +| 251 | blk.27.ffn_down.weight | 0x174b24500 | 0x44c0000 | +| 252 | blk.27.ffn_gate.weight | 0x178fe4500 | 0x44c0000 | +| 253 | blk.27.ffn_norm.weight | 0x17d4a4500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x17d4a9500 | 0x44c0000 | +| 255 | blk.28.attn_k.weight | 0x181969500 | 0x226000 | +| 256 | blk.28.attn_norm.weight | 0x181b8f500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x181b94500 | 0x898000 | +| 258 | blk.28.attn_q.weight | 0x18242c500 | 0x898000 | +| 259 | blk.28.attn_v.weight | 0x182cc4500 | 0x226000 | +| 260 | blk.28.ffn_down.weight | 0x182eea500 | 0x44c0000 | +| 261 | blk.28.ffn_gate.weight | 0x1873aa500 | 0x44c0000 | +| 262 | blk.28.ffn_norm.weight | 0x18b86a500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x18b86f500 | 0x44c0000 | +| 264 | blk.29.attn_k.weight | 0x18fd2f500 | 0x226000 | +| 265 | blk.29.attn_norm.weight | 0x18ff55500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x18ff5a500 | 0x898000 | +| 267 | blk.29.attn_q.weight | 0x1907f2500 | 0x898000 | +| 268 | blk.29.attn_v.weight | 0x19108a500 | 0x226000 | +| 269 | blk.29.ffn_down.weight | 0x1912b0500 | 0x44c0000 | +| 270 | blk.29.ffn_gate.weight | 0x195770500 | 0x44c0000 | +| 271 | blk.29.ffn_norm.weight | 0x199c30500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x199c35500 | 0x44c0000 | +| 273 | blk.30.attn_k.weight | 0x19e0f5500 | 0x226000 | +| 274 | blk.30.attn_norm.weight | 0x19e31b500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x19e320500 | 0x898000 | +| 276 | blk.30.attn_q.weight | 0x19ebb8500 | 0x898000 | +| 277 | blk.30.attn_v.weight | 0x19f450500 | 0x226000 | +| 278 | blk.30.ffn_down.weight | 0x19f676500 | 0x44c0000 | +| 279 | blk.30.ffn_gate.weight | 0x1a3b36500 | 0x44c0000 | +| 280 | blk.30.ffn_norm.weight | 0x1a7ff6500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x1a7ffb500 | 0x44c0000 | +| 282 | blk.31.attn_k.weight | 0x1ac4bb500 | 0x226000 | +| 283 | blk.31.attn_norm.weight | 0x1ac6e1500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x1ac6e6500 | 0x898000 | +| 285 | blk.31.attn_q.weight | 0x1acf7e500 | 0x898000 | +| 286 | blk.31.attn_v.weight | 0x1ad816500 | 0x226000 | +| 287 | blk.31.ffn_down.weight | 0x1ada3c500 | 0x44c0000 | +| 288 | blk.31.ffn_gate.weight | 0x1b1efc500 | 0x44c0000 | +| 289 | blk.31.ffn_norm.weight | 0x1b63bc500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x1b63c1500 | 0x44c0000 | +| 291 | blk.32.attn_k.weight | 0x1ba881500 | 0x226000 | +| 292 | blk.32.attn_norm.weight | 0x1baaa7500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x1baaac500 | 0x898000 | +| 294 | blk.32.attn_q.weight | 0x1bb344500 | 0x898000 | +| 295 | blk.32.attn_v.weight | 0x1bbbdc500 | 0x226000 | +| 296 | blk.32.ffn_down.weight | 0x1bbe02500 | 0x44c0000 | +| 297 | blk.32.ffn_gate.weight | 0x1c02c2500 | 0x44c0000 | +| 298 | blk.32.ffn_norm.weight | 0x1c4782500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x1c4787500 | 0x44c0000 | +| 300 | blk.33.attn_k.weight | 0x1c8c47500 | 0x226000 | +| 301 | blk.33.attn_norm.weight | 0x1c8e6d500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x1c8e72500 | 0x898000 | +| 303 | blk.33.attn_q.weight | 0x1c970a500 | 0x898000 | +| 304 | blk.33.attn_v.weight | 0x1c9fa2500 | 0x226000 | +| 305 | blk.33.ffn_down.weight | 0x1ca1c8500 | 0x44c0000 | +| 306 | blk.33.ffn_gate.weight | 0x1ce688500 | 0x44c0000 | +| 307 | blk.33.ffn_norm.weight | 0x1d2b48500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x1d2b4d500 | 0x44c0000 | +| 309 | blk.34.attn_k.weight | 0x1d700d500 | 0x226000 | +| 310 | blk.34.attn_norm.weight | 0x1d7233500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x1d7238500 | 0x898000 | +| 312 | blk.34.attn_q.weight | 0x1d7ad0500 | 0x898000 | +| 313 | blk.34.attn_v.weight | 0x1d8368500 | 0x226000 | +| 314 | blk.34.ffn_down.weight | 0x1d858e500 | 0x44c0000 | +| 315 | blk.34.ffn_gate.weight | 0x1dca4e500 | 0x44c0000 | +| 316 | blk.34.ffn_norm.weight | 0x1e0f0e500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x1e0f13500 | 0x44c0000 | +| 318 | blk.35.attn_k.weight | 0x1e53d3500 | 0x226000 | +| 319 | blk.35.attn_norm.weight | 0x1e55f9500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x1e55fe500 | 0x898000 | +| 321 | blk.35.attn_q.weight | 0x1e5e96500 | 0x898000 | +| 322 | blk.35.attn_v.weight | 0x1e672e500 | 0x226000 | +| 323 | blk.35.ffn_down.weight | 0x1e6954500 | 0x44c0000 | +| 324 | blk.35.ffn_gate.weight | 0x1eae14500 | 0x44c0000 | +| 325 | blk.35.ffn_norm.weight | 0x1ef2d4500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x1ef2d9500 | 0x44c0000 | +| 327 | blk.36.attn_k.weight | 0x1f3799500 | 0x226000 | +| 328 | blk.36.attn_norm.weight | 0x1f39bf500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x1f39c4500 | 0x898000 | +| 330 | blk.36.attn_q.weight | 0x1f425c500 | 0x898000 | +| 331 | blk.36.attn_v.weight | 0x1f4af4500 | 0x226000 | +| 332 | blk.36.ffn_down.weight | 0x1f4d1a500 | 0x44c0000 | +| 333 | blk.36.ffn_gate.weight | 0x1f91da500 | 0x44c0000 | +| 334 | blk.36.ffn_norm.weight | 0x1fd69a500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x1fd69f500 | 0x44c0000 | +| 336 | blk.37.attn_k.weight | 0x201b5f500 | 0x226000 | +| 337 | blk.37.attn_norm.weight | 0x201d85500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x201d8a500 | 0x898000 | +| 339 | blk.37.attn_q.weight | 0x202622500 | 0x898000 | +| 340 | blk.37.attn_v.weight | 0x202eba500 | 0x226000 | +| 341 | blk.37.ffn_down.weight | 0x2030e0500 | 0x44c0000 | +| 342 | blk.37.ffn_gate.weight | 0x2075a0500 | 0x44c0000 | +| 343 | blk.37.ffn_norm.weight | 0x20ba60500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x20ba65500 | 0x44c0000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q2_K | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q3_K | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q3_K | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-Q4_K_M.md b/scores/Dolphin-Mistral-24B-Venice-Edition-Q4_K_M.md new file mode 100644 index 0000000..ecd2a65 --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-Q4_K_M.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-Q4_K_M.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 15 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-Q4\_K\_M.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-q4_k_mgguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0x16800000 | +| 1 | output_norm.weight | 0x16f84500 | 0x5000 | +| 2 | token_embd.weight | 0x16f89500 | 0x11300000 | +| 3 | blk.0.attn_k.weight | 0x28289500 | 0x226000 | +| 4 | blk.0.attn_norm.weight | 0x284af500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x284b4500 | 0xb40000 | +| 6 | blk.0.attn_q.weight | 0x28ff4500 | 0x898000 | +| 7 | blk.0.attn_v.weight | 0x2988c500 | 0x2d0000 | +| 8 | blk.0.ffn_down.weight | 0x29b5c500 | 0x8340000 | +| 9 | blk.0.ffn_gate.weight | 0x31e9c500 | 0x44c0000 | +| 10 | blk.0.ffn_norm.weight | 0x3635c500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x36361500 | 0x44c0000 | +| 12 | blk.1.attn_k.weight | 0x3a821500 | 0x226000 | +| 13 | blk.1.attn_norm.weight | 0x3aa47500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x3aa4c500 | 0xb40000 | +| 15 | blk.1.attn_q.weight | 0x3b58c500 | 0x898000 | +| 16 | blk.1.attn_v.weight | 0x3be24500 | 0x2d0000 | +| 17 | blk.1.ffn_down.weight | 0x3c0f4500 | 0x8340000 | +| 18 | blk.1.ffn_gate.weight | 0x44434500 | 0x44c0000 | +| 19 | blk.1.ffn_norm.weight | 0x488f4500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x488f9500 | 0x44c0000 | +| 21 | blk.2.attn_k.weight | 0x4cdb9500 | 0x226000 | +| 22 | blk.2.attn_norm.weight | 0x4cfdf500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x4cfe4500 | 0xb40000 | +| 24 | blk.2.attn_q.weight | 0x4db24500 | 0x898000 | +| 25 | blk.2.attn_v.weight | 0x4e3bc500 | 0x2d0000 | +| 26 | blk.2.ffn_down.weight | 0x4e68c500 | 0x8340000 | +| 27 | blk.2.ffn_gate.weight | 0x569cc500 | 0x44c0000 | +| 28 | blk.2.ffn_norm.weight | 0x5ae8c500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x5ae91500 | 0x44c0000 | +| 30 | blk.3.attn_k.weight | 0x5f351500 | 0x226000 | +| 31 | blk.3.attn_norm.weight | 0x5f577500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x5f57c500 | 0xb40000 | +| 33 | blk.3.attn_q.weight | 0x600bc500 | 0x898000 | +| 34 | blk.3.attn_v.weight | 0x60954500 | 0x2d0000 | +| 35 | blk.3.ffn_down.weight | 0x60c24500 | 0x8340000 | +| 36 | blk.3.ffn_gate.weight | 0x68f64500 | 0x44c0000 | +| 37 | blk.3.ffn_norm.weight | 0x6d424500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0x6d429500 | 0x44c0000 | +| 39 | blk.4.attn_k.weight | 0x718e9500 | 0x226000 | +| 40 | blk.4.attn_norm.weight | 0x71b0f500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0x71b14500 | 0xb40000 | +| 42 | blk.4.attn_q.weight | 0x72654500 | 0x898000 | +| 43 | blk.4.attn_v.weight | 0x72eec500 | 0x370000 | +| 44 | blk.4.ffn_down.weight | 0x7325c500 | 0x8340000 | +| 45 | blk.4.ffn_gate.weight | 0x7b59c500 | 0x44c0000 | +| 46 | blk.4.ffn_norm.weight | 0x7fa5c500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0x7fa61500 | 0x44c0000 | +| 48 | blk.5.attn_k.weight | 0x83f21500 | 0x226000 | +| 49 | blk.5.attn_norm.weight | 0x84147500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0x8414c500 | 0xb40000 | +| 51 | blk.5.attn_q.weight | 0x84c8c500 | 0x898000 | +| 52 | blk.5.attn_v.weight | 0x85524500 | 0x370000 | +| 53 | blk.5.ffn_down.weight | 0x85894500 | 0x5a00000 | +| 54 | blk.5.ffn_gate.weight | 0x8b294500 | 0x44c0000 | +| 55 | blk.5.ffn_norm.weight | 0x8f754500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0x8f759500 | 0x44c0000 | +| 57 | blk.6.attn_k.weight | 0x93c19500 | 0x226000 | +| 58 | blk.6.attn_norm.weight | 0x93e3f500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0x93e44500 | 0xb40000 | +| 60 | blk.6.attn_q.weight | 0x94984500 | 0x898000 | +| 61 | blk.6.attn_v.weight | 0x9521c500 | 0x2d0000 | +| 62 | blk.6.ffn_down.weight | 0x954ec500 | 0x5a00000 | +| 63 | blk.6.ffn_gate.weight | 0x9aeec500 | 0x44c0000 | +| 64 | blk.6.ffn_norm.weight | 0x9f3ac500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0x9f3b1500 | 0x44c0000 | +| 66 | blk.7.attn_k.weight | 0xa3871500 | 0x226000 | +| 67 | blk.7.attn_norm.weight | 0xa3a97500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0xa3a9c500 | 0xb40000 | +| 69 | blk.7.attn_q.weight | 0xa45dc500 | 0x898000 | +| 70 | blk.7.attn_v.weight | 0xa4e74500 | 0x370000 | +| 71 | blk.7.ffn_down.weight | 0xa51e4500 | 0x8340000 | +| 72 | blk.7.ffn_gate.weight | 0xad524500 | 0x44c0000 | +| 73 | blk.7.ffn_norm.weight | 0xb19e4500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0xb19e9500 | 0x44c0000 | +| 75 | blk.8.attn_k.weight | 0xb5ea9500 | 0x226000 | +| 76 | blk.8.attn_norm.weight | 0xb60cf500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0xb60d4500 | 0xb40000 | +| 78 | blk.8.attn_q.weight | 0xb6c14500 | 0x898000 | +| 79 | blk.8.attn_v.weight | 0xb74ac500 | 0x370000 | +| 80 | blk.8.ffn_down.weight | 0xb781c500 | 0x5a00000 | +| 81 | blk.8.ffn_gate.weight | 0xbd21c500 | 0x44c0000 | +| 82 | blk.8.ffn_norm.weight | 0xc16dc500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0xc16e1500 | 0x44c0000 | +| 84 | blk.9.attn_k.weight | 0xc5ba1500 | 0x226000 | +| 85 | blk.9.attn_norm.weight | 0xc5dc7500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0xc5dcc500 | 0xb40000 | +| 87 | blk.9.attn_q.weight | 0xc690c500 | 0x898000 | +| 88 | blk.9.attn_v.weight | 0xc71a4500 | 0x2d0000 | +| 89 | blk.9.ffn_down.weight | 0xc7474500 | 0x5a00000 | +| 90 | blk.9.ffn_gate.weight | 0xcce74500 | 0x44c0000 | +| 91 | blk.9.ffn_norm.weight | 0xd1334500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0xd1339500 | 0x44c0000 | +| 93 | blk.10.attn_k.weight | 0xd57f9500 | 0x226000 | +| 94 | blk.10.attn_norm.weight | 0xd5a1f500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0xd5a24500 | 0xb40000 | +| 96 | blk.10.attn_q.weight | 0xd6564500 | 0x898000 | +| 97 | blk.10.attn_v.weight | 0xd6dfc500 | 0x370000 | +| 98 | blk.10.ffn_down.weight | 0xd716c500 | 0x8340000 | +| 99 | blk.10.ffn_gate.weight | 0xdf4ac500 | 0x44c0000 | +| 100 | blk.10.ffn_norm.weight | 0xe396c500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0xe3971500 | 0x44c0000 | +| 102 | blk.11.attn_k.weight | 0xe7e31500 | 0x226000 | +| 103 | blk.11.attn_norm.weight | 0xe8057500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0xe805c500 | 0xb40000 | +| 105 | blk.11.attn_q.weight | 0xe8b9c500 | 0x898000 | +| 106 | blk.11.attn_v.weight | 0xe9434500 | 0x370000 | +| 107 | blk.11.ffn_down.weight | 0xe97a4500 | 0x5a00000 | +| 108 | blk.11.ffn_gate.weight | 0xef1a4500 | 0x44c0000 | +| 109 | blk.11.ffn_norm.weight | 0xf3664500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0xf3669500 | 0x44c0000 | +| 111 | blk.12.attn_k.weight | 0xf7b29500 | 0x226000 | +| 112 | blk.12.attn_norm.weight | 0xf7d4f500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0xf7d54500 | 0xb40000 | +| 114 | blk.12.attn_q.weight | 0xf8894500 | 0x898000 | +| 115 | blk.12.attn_v.weight | 0xf912c500 | 0x2d0000 | +| 116 | blk.12.ffn_down.weight | 0xf93fc500 | 0x5a00000 | +| 117 | blk.12.ffn_gate.weight | 0xfedfc500 | 0x44c0000 | +| 118 | blk.12.ffn_norm.weight | 0x1032bc500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0x1032c1500 | 0x44c0000 | +| 120 | blk.13.attn_k.weight | 0x107781500 | 0x226000 | +| 121 | blk.13.attn_norm.weight | 0x1079a7500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0x1079ac500 | 0xb40000 | +| 123 | blk.13.attn_q.weight | 0x1084ec500 | 0x898000 | +| 124 | blk.13.attn_v.weight | 0x108d84500 | 0x370000 | +| 125 | blk.13.ffn_down.weight | 0x1090f4500 | 0x8340000 | +| 126 | blk.13.ffn_gate.weight | 0x111434500 | 0x44c0000 | +| 127 | blk.13.ffn_norm.weight | 0x1158f4500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0x1158f9500 | 0x44c0000 | +| 129 | blk.14.attn_k.weight | 0x119db9500 | 0x226000 | +| 130 | blk.14.attn_norm.weight | 0x119fdf500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0x119fe4500 | 0xb40000 | +| 132 | blk.14.attn_q.weight | 0x11ab24500 | 0x898000 | +| 133 | blk.14.attn_v.weight | 0x11b3bc500 | 0x370000 | +| 134 | blk.14.ffn_down.weight | 0x11b72c500 | 0x5a00000 | +| 135 | blk.14.ffn_gate.weight | 0x12112c500 | 0x44c0000 | +| 136 | blk.14.ffn_norm.weight | 0x1255ec500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0x1255f1500 | 0x44c0000 | +| 138 | blk.15.attn_k.weight | 0x129ab1500 | 0x226000 | +| 139 | blk.15.attn_norm.weight | 0x129cd7500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0x129cdc500 | 0xb40000 | +| 141 | blk.15.attn_q.weight | 0x12a81c500 | 0x898000 | +| 142 | blk.15.attn_v.weight | 0x12b0b4500 | 0x2d0000 | +| 143 | blk.15.ffn_down.weight | 0x12b384500 | 0x5a00000 | +| 144 | blk.15.ffn_gate.weight | 0x130d84500 | 0x44c0000 | +| 145 | blk.15.ffn_norm.weight | 0x135244500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0x135249500 | 0x44c0000 | +| 147 | blk.16.attn_k.weight | 0x139709500 | 0x226000 | +| 148 | blk.16.attn_norm.weight | 0x13992f500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0x139934500 | 0xb40000 | +| 150 | blk.16.attn_q.weight | 0x13a474500 | 0x898000 | +| 151 | blk.16.attn_v.weight | 0x13ad0c500 | 0x370000 | +| 152 | blk.16.ffn_down.weight | 0x13b07c500 | 0x8340000 | +| 153 | blk.16.ffn_gate.weight | 0x1433bc500 | 0x44c0000 | +| 154 | blk.16.ffn_norm.weight | 0x14787c500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0x147881500 | 0x44c0000 | +| 156 | blk.17.attn_k.weight | 0x14bd41500 | 0x2d0000 | +| 157 | blk.17.attn_norm.weight | 0x14c011500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0x14c016500 | 0xb40000 | +| 159 | blk.17.attn_q.weight | 0x14cb56500 | 0xb40000 | +| 160 | blk.17.attn_v.weight | 0x14d696500 | 0x370000 | +| 161 | blk.17.ffn_down.weight | 0x14da06500 | 0x5a00000 | +| 162 | blk.17.ffn_gate.weight | 0x153406500 | 0x44c0000 | +| 163 | blk.17.ffn_norm.weight | 0x1578c6500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0x1578cb500 | 0x44c0000 | +| 165 | blk.18.attn_k.weight | 0x15bd8b500 | 0x2d0000 | +| 166 | blk.18.attn_norm.weight | 0x15c05b500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0x15c060500 | 0xb40000 | +| 168 | blk.18.attn_q.weight | 0x15cba0500 | 0xb40000 | +| 169 | blk.18.attn_v.weight | 0x15d6e0500 | 0x370000 | +| 170 | blk.18.ffn_down.weight | 0x15da50500 | 0x5a00000 | +| 171 | blk.18.ffn_gate.weight | 0x163450500 | 0x44c0000 | +| 172 | blk.18.ffn_norm.weight | 0x167910500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x167915500 | 0x44c0000 | +| 174 | blk.19.attn_k.weight | 0x16bdd5500 | 0x226000 | +| 175 | blk.19.attn_norm.weight | 0x16bffb500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x16c000500 | 0xb40000 | +| 177 | blk.19.attn_q.weight | 0x16cb40500 | 0x898000 | +| 178 | blk.19.attn_v.weight | 0x16d3d8500 | 0x370000 | +| 179 | blk.19.ffn_down.weight | 0x16d748500 | 0x8340000 | +| 180 | blk.19.ffn_gate.weight | 0x175a88500 | 0x44c0000 | +| 181 | blk.19.ffn_norm.weight | 0x179f48500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x179f4d500 | 0x44c0000 | +| 183 | blk.20.attn_k.weight | 0x17e40d500 | 0x2d0000 | +| 184 | blk.20.attn_norm.weight | 0x17e6dd500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x17e6e2500 | 0xb40000 | +| 186 | blk.20.attn_q.weight | 0x17f222500 | 0xb40000 | +| 187 | blk.20.attn_v.weight | 0x17fd62500 | 0x370000 | +| 188 | blk.20.ffn_down.weight | 0x1800d2500 | 0x5a00000 | +| 189 | blk.20.ffn_gate.weight | 0x185ad2500 | 0x5a00000 | +| 190 | blk.20.ffn_norm.weight | 0x18b4d2500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x18b4d7500 | 0x5a00000 | +| 192 | blk.21.attn_k.weight | 0x190ed7500 | 0x226000 | +| 193 | blk.21.attn_norm.weight | 0x1910fd500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x191102500 | 0xb40000 | +| 195 | blk.21.attn_q.weight | 0x191c42500 | 0x898000 | +| 196 | blk.21.attn_v.weight | 0x1924da500 | 0x2d0000 | +| 197 | blk.21.ffn_down.weight | 0x1927aa500 | 0x5a00000 | +| 198 | blk.21.ffn_gate.weight | 0x1981aa500 | 0x5a00000 | +| 199 | blk.21.ffn_norm.weight | 0x19dbaa500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x19dbaf500 | 0x5a00000 | +| 201 | blk.22.attn_k.weight | 0x1a35af500 | 0x2d0000 | +| 202 | blk.22.attn_norm.weight | 0x1a387f500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x1a3884500 | 0xb40000 | +| 204 | blk.22.attn_q.weight | 0x1a43c4500 | 0xb40000 | +| 205 | blk.22.attn_v.weight | 0x1a4f04500 | 0x370000 | +| 206 | blk.22.ffn_down.weight | 0x1a5274500 | 0x8340000 | +| 207 | blk.22.ffn_gate.weight | 0x1ad5b4500 | 0x5a00000 | +| 208 | blk.22.ffn_norm.weight | 0x1b2fb4500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x1b2fb9500 | 0x5a00000 | +| 210 | blk.23.attn_k.weight | 0x1b89b9500 | 0x2d0000 | +| 211 | blk.23.attn_norm.weight | 0x1b8c89500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x1b8c8e500 | 0xb40000 | +| 213 | blk.23.attn_q.weight | 0x1b97ce500 | 0xb40000 | +| 214 | blk.23.attn_v.weight | 0x1ba30e500 | 0x370000 | +| 215 | blk.23.ffn_down.weight | 0x1ba67e500 | 0x5a00000 | +| 216 | blk.23.ffn_gate.weight | 0x1c007e500 | 0x5a00000 | +| 217 | blk.23.ffn_norm.weight | 0x1c5a7e500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x1c5a83500 | 0x5a00000 | +| 219 | blk.24.attn_k.weight | 0x1cb483500 | 0x2d0000 | +| 220 | blk.24.attn_norm.weight | 0x1cb753500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x1cb758500 | 0xb40000 | +| 222 | blk.24.attn_q.weight | 0x1cc298500 | 0xb40000 | +| 223 | blk.24.attn_v.weight | 0x1ccdd8500 | 0x370000 | +| 224 | blk.24.ffn_down.weight | 0x1cd148500 | 0x5a00000 | +| 225 | blk.24.ffn_gate.weight | 0x1d2b48500 | 0x5a00000 | +| 226 | blk.24.ffn_norm.weight | 0x1d8548500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x1d854d500 | 0x5a00000 | +| 228 | blk.25.attn_k.weight | 0x1ddf4d500 | 0x2d0000 | +| 229 | blk.25.attn_norm.weight | 0x1de21d500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x1de222500 | 0xb40000 | +| 231 | blk.25.attn_q.weight | 0x1ded62500 | 0xb40000 | +| 232 | blk.25.attn_v.weight | 0x1df8a2500 | 0x370000 | +| 233 | blk.25.ffn_down.weight | 0x1dfc12500 | 0x8340000 | +| 234 | blk.25.ffn_gate.weight | 0x1e7f52500 | 0x5a00000 | +| 235 | blk.25.ffn_norm.weight | 0x1ed952500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x1ed957500 | 0x5a00000 | +| 237 | blk.26.attn_k.weight | 0x1f3357500 | 0x2d0000 | +| 238 | blk.26.attn_norm.weight | 0x1f3627500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x1f362c500 | 0xb40000 | +| 240 | blk.26.attn_q.weight | 0x1f416c500 | 0xb40000 | +| 241 | blk.26.attn_v.weight | 0x1f4cac500 | 0x370000 | +| 242 | blk.26.ffn_down.weight | 0x1f501c500 | 0x5a00000 | +| 243 | blk.26.ffn_gate.weight | 0x1faa1c500 | 0x5a00000 | +| 244 | blk.26.ffn_norm.weight | 0x20041c500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x200421500 | 0x5a00000 | +| 246 | blk.27.attn_k.weight | 0x205e21500 | 0x226000 | +| 247 | blk.27.attn_norm.weight | 0x206047500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x20604c500 | 0xb40000 | +| 249 | blk.27.attn_q.weight | 0x206b8c500 | 0x898000 | +| 250 | blk.27.attn_v.weight | 0x207424500 | 0x2d0000 | +| 251 | blk.27.ffn_down.weight | 0x2076f4500 | 0x5a00000 | +| 252 | blk.27.ffn_gate.weight | 0x20d0f4500 | 0x5a00000 | +| 253 | blk.27.ffn_norm.weight | 0x212af4500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x212af9500 | 0x5a00000 | +| 255 | blk.28.attn_k.weight | 0x2184f9500 | 0x2d0000 | +| 256 | blk.28.attn_norm.weight | 0x2187c9500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x2187ce500 | 0xb40000 | +| 258 | blk.28.attn_q.weight | 0x21930e500 | 0xb40000 | +| 259 | blk.28.attn_v.weight | 0x219e4e500 | 0x370000 | +| 260 | blk.28.ffn_down.weight | 0x21a1be500 | 0x8340000 | +| 261 | blk.28.ffn_gate.weight | 0x2224fe500 | 0x5a00000 | +| 262 | blk.28.ffn_norm.weight | 0x227efe500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x227f03500 | 0x5a00000 | +| 264 | blk.29.attn_k.weight | 0x22d903500 | 0x2d0000 | +| 265 | blk.29.attn_norm.weight | 0x22dbd3500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x22dbd8500 | 0xb40000 | +| 267 | blk.29.attn_q.weight | 0x22e718500 | 0xb40000 | +| 268 | blk.29.attn_v.weight | 0x22f258500 | 0x370000 | +| 269 | blk.29.ffn_down.weight | 0x22f5c8500 | 0x5a00000 | +| 270 | blk.29.ffn_gate.weight | 0x234fc8500 | 0x5a00000 | +| 271 | blk.29.ffn_norm.weight | 0x23a9c8500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x23a9cd500 | 0x5a00000 | +| 273 | blk.30.attn_k.weight | 0x2403cd500 | 0x2d0000 | +| 274 | blk.30.attn_norm.weight | 0x24069d500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x2406a2500 | 0xb40000 | +| 276 | blk.30.attn_q.weight | 0x2411e2500 | 0xb40000 | +| 277 | blk.30.attn_v.weight | 0x241d22500 | 0x370000 | +| 278 | blk.30.ffn_down.weight | 0x242092500 | 0x5a00000 | +| 279 | blk.30.ffn_gate.weight | 0x247a92500 | 0x5a00000 | +| 280 | blk.30.ffn_norm.weight | 0x24d492500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x24d497500 | 0x5a00000 | +| 282 | blk.31.attn_k.weight | 0x252e97500 | 0x2d0000 | +| 283 | blk.31.attn_norm.weight | 0x253167500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x25316c500 | 0xb40000 | +| 285 | blk.31.attn_q.weight | 0x253cac500 | 0xb40000 | +| 286 | blk.31.attn_v.weight | 0x2547ec500 | 0x370000 | +| 287 | blk.31.ffn_down.weight | 0x254b5c500 | 0x8340000 | +| 288 | blk.31.ffn_gate.weight | 0x25ce9c500 | 0x5a00000 | +| 289 | blk.31.ffn_norm.weight | 0x26289c500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x2628a1500 | 0x5a00000 | +| 291 | blk.32.attn_k.weight | 0x2682a1500 | 0x2d0000 | +| 292 | blk.32.attn_norm.weight | 0x268571500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x268576500 | 0xb40000 | +| 294 | blk.32.attn_q.weight | 0x2690b6500 | 0xb40000 | +| 295 | blk.32.attn_v.weight | 0x269bf6500 | 0x370000 | +| 296 | blk.32.ffn_down.weight | 0x269f66500 | 0x5a00000 | +| 297 | blk.32.ffn_gate.weight | 0x26f966500 | 0x5a00000 | +| 298 | blk.32.ffn_norm.weight | 0x275366500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x27536b500 | 0x5a00000 | +| 300 | blk.33.attn_k.weight | 0x27ad6b500 | 0x2d0000 | +| 301 | blk.33.attn_norm.weight | 0x27b03b500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x27b040500 | 0xb40000 | +| 303 | blk.33.attn_q.weight | 0x27bb80500 | 0xb40000 | +| 304 | blk.33.attn_v.weight | 0x27c6c0500 | 0x370000 | +| 305 | blk.33.ffn_down.weight | 0x27ca30500 | 0x5a00000 | +| 306 | blk.33.ffn_gate.weight | 0x282430500 | 0x5a00000 | +| 307 | blk.33.ffn_norm.weight | 0x287e30500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x287e35500 | 0x5a00000 | +| 309 | blk.34.attn_k.weight | 0x28d835500 | 0x2d0000 | +| 310 | blk.34.attn_norm.weight | 0x28db05500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x28db0a500 | 0xb40000 | +| 312 | blk.34.attn_q.weight | 0x28e64a500 | 0xb40000 | +| 313 | blk.34.attn_v.weight | 0x28f18a500 | 0x370000 | +| 314 | blk.34.ffn_down.weight | 0x28f4fa500 | 0x8340000 | +| 315 | blk.34.ffn_gate.weight | 0x29783a500 | 0x5a00000 | +| 316 | blk.34.ffn_norm.weight | 0x29d23a500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x29d23f500 | 0x5a00000 | +| 318 | blk.35.attn_k.weight | 0x2a2c3f500 | 0x2d0000 | +| 319 | blk.35.attn_norm.weight | 0x2a2f0f500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x2a2f14500 | 0xb40000 | +| 321 | blk.35.attn_q.weight | 0x2a3a54500 | 0xb40000 | +| 322 | blk.35.attn_v.weight | 0x2a4594500 | 0x370000 | +| 323 | blk.35.ffn_down.weight | 0x2a4904500 | 0x8340000 | +| 324 | blk.35.ffn_gate.weight | 0x2acc44500 | 0x5a00000 | +| 325 | blk.35.ffn_norm.weight | 0x2b2644500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x2b2649500 | 0x5a00000 | +| 327 | blk.36.attn_k.weight | 0x2b8049500 | 0x2d0000 | +| 328 | blk.36.attn_norm.weight | 0x2b8319500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x2b831e500 | 0xb40000 | +| 330 | blk.36.attn_q.weight | 0x2b8e5e500 | 0xb40000 | +| 331 | blk.36.attn_v.weight | 0x2b999e500 | 0x370000 | +| 332 | blk.36.ffn_down.weight | 0x2b9d0e500 | 0x8340000 | +| 333 | blk.36.ffn_gate.weight | 0x2c204e500 | 0x5a00000 | +| 334 | blk.36.ffn_norm.weight | 0x2c7a4e500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x2c7a53500 | 0x5a00000 | +| 336 | blk.37.attn_k.weight | 0x2cd453500 | 0x2d0000 | +| 337 | blk.37.attn_norm.weight | 0x2cd723500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x2cd728500 | 0xb40000 | +| 339 | blk.37.attn_q.weight | 0x2ce268500 | 0xb40000 | +| 340 | blk.37.attn_v.weight | 0x2ceda8500 | 0x370000 | +| 341 | blk.37.ffn_down.weight | 0x2cf118500 | 0x8340000 | +| 342 | blk.37.ffn_gate.weight | 0x2d7458500 | 0x5a00000 | +| 343 | blk.37.ffn_norm.weight | 0x2dce58500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x2dce5d500 | 0x5a00000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q4_K | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-Q4_K_S.md b/scores/Dolphin-Mistral-24B-Venice-Edition-Q4_K_S.md new file mode 100644 index 0000000..87b0d4f --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-Q4_K_S.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-Q4_K_S.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 14 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-Q4\_K\_S.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-q4_k_sgguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0x16800000 | +| 1 | output_norm.weight | 0x16f84500 | 0x5000 | +| 2 | token_embd.weight | 0x16f89500 | 0x11300000 | +| 3 | blk.0.attn_k.weight | 0x28289500 | 0x226000 | +| 4 | blk.0.attn_norm.weight | 0x284af500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x284b4500 | 0xb40000 | +| 6 | blk.0.attn_q.weight | 0x28ff4500 | 0x898000 | +| 7 | blk.0.attn_v.weight | 0x2988c500 | 0x2d0000 | +| 8 | blk.0.ffn_down.weight | 0x29b5c500 | 0x6e00000 | +| 9 | blk.0.ffn_gate.weight | 0x3095c500 | 0x44c0000 | +| 10 | blk.0.ffn_norm.weight | 0x34e1c500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x34e21500 | 0x44c0000 | +| 12 | blk.1.attn_k.weight | 0x392e1500 | 0x226000 | +| 13 | blk.1.attn_norm.weight | 0x39507500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x3950c500 | 0xb40000 | +| 15 | blk.1.attn_q.weight | 0x3a04c500 | 0x898000 | +| 16 | blk.1.attn_v.weight | 0x3a8e4500 | 0x2d0000 | +| 17 | blk.1.ffn_down.weight | 0x3abb4500 | 0x6e00000 | +| 18 | blk.1.ffn_gate.weight | 0x419b4500 | 0x44c0000 | +| 19 | blk.1.ffn_norm.weight | 0x45e74500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x45e79500 | 0x44c0000 | +| 21 | blk.2.attn_k.weight | 0x4a339500 | 0x226000 | +| 22 | blk.2.attn_norm.weight | 0x4a55f500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x4a564500 | 0xb40000 | +| 24 | blk.2.attn_q.weight | 0x4b0a4500 | 0x898000 | +| 25 | blk.2.attn_v.weight | 0x4b93c500 | 0x2d0000 | +| 26 | blk.2.ffn_down.weight | 0x4bc0c500 | 0x6e00000 | +| 27 | blk.2.ffn_gate.weight | 0x52a0c500 | 0x44c0000 | +| 28 | blk.2.ffn_norm.weight | 0x56ecc500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x56ed1500 | 0x44c0000 | +| 30 | blk.3.attn_k.weight | 0x5b391500 | 0x226000 | +| 31 | blk.3.attn_norm.weight | 0x5b5b7500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x5b5bc500 | 0xb40000 | +| 33 | blk.3.attn_q.weight | 0x5c0fc500 | 0x898000 | +| 34 | blk.3.attn_v.weight | 0x5c994500 | 0x2d0000 | +| 35 | blk.3.ffn_down.weight | 0x5cc64500 | 0x6e00000 | +| 36 | blk.3.ffn_gate.weight | 0x63a64500 | 0x44c0000 | +| 37 | blk.3.ffn_norm.weight | 0x67f24500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0x67f29500 | 0x44c0000 | +| 39 | blk.4.attn_k.weight | 0x6c3e9500 | 0x226000 | +| 40 | blk.4.attn_norm.weight | 0x6c60f500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0x6c614500 | 0xb40000 | +| 42 | blk.4.attn_q.weight | 0x6d154500 | 0x898000 | +| 43 | blk.4.attn_v.weight | 0x6d9ec500 | 0x2d0000 | +| 44 | blk.4.ffn_down.weight | 0x6dcbc500 | 0x6e00000 | +| 45 | blk.4.ffn_gate.weight | 0x74abc500 | 0x44c0000 | +| 46 | blk.4.ffn_norm.weight | 0x78f7c500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0x78f81500 | 0x44c0000 | +| 48 | blk.5.attn_k.weight | 0x7d441500 | 0x226000 | +| 49 | blk.5.attn_norm.weight | 0x7d667500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0x7d66c500 | 0xb40000 | +| 51 | blk.5.attn_q.weight | 0x7e1ac500 | 0x898000 | +| 52 | blk.5.attn_v.weight | 0x7ea44500 | 0x2d0000 | +| 53 | blk.5.ffn_down.weight | 0x7ed14500 | 0x5a00000 | +| 54 | blk.5.ffn_gate.weight | 0x84714500 | 0x44c0000 | +| 55 | blk.5.ffn_norm.weight | 0x88bd4500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0x88bd9500 | 0x44c0000 | +| 57 | blk.6.attn_k.weight | 0x8d099500 | 0x226000 | +| 58 | blk.6.attn_norm.weight | 0x8d2bf500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0x8d2c4500 | 0xb40000 | +| 60 | blk.6.attn_q.weight | 0x8de04500 | 0x898000 | +| 61 | blk.6.attn_v.weight | 0x8e69c500 | 0x2d0000 | +| 62 | blk.6.ffn_down.weight | 0x8e96c500 | 0x5a00000 | +| 63 | blk.6.ffn_gate.weight | 0x9436c500 | 0x44c0000 | +| 64 | blk.6.ffn_norm.weight | 0x9882c500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0x98831500 | 0x44c0000 | +| 66 | blk.7.attn_k.weight | 0x9ccf1500 | 0x226000 | +| 67 | blk.7.attn_norm.weight | 0x9cf17500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0x9cf1c500 | 0xb40000 | +| 69 | blk.7.attn_q.weight | 0x9da5c500 | 0x898000 | +| 70 | blk.7.attn_v.weight | 0x9e2f4500 | 0x2d0000 | +| 71 | blk.7.ffn_down.weight | 0x9e5c4500 | 0x5a00000 | +| 72 | blk.7.ffn_gate.weight | 0xa3fc4500 | 0x44c0000 | +| 73 | blk.7.ffn_norm.weight | 0xa8484500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0xa8489500 | 0x44c0000 | +| 75 | blk.8.attn_k.weight | 0xac949500 | 0x226000 | +| 76 | blk.8.attn_norm.weight | 0xacb6f500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0xacb74500 | 0xb40000 | +| 78 | blk.8.attn_q.weight | 0xad6b4500 | 0x898000 | +| 79 | blk.8.attn_v.weight | 0xadf4c500 | 0x2d0000 | +| 80 | blk.8.ffn_down.weight | 0xae21c500 | 0x5a00000 | +| 81 | blk.8.ffn_gate.weight | 0xb3c1c500 | 0x44c0000 | +| 82 | blk.8.ffn_norm.weight | 0xb80dc500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0xb80e1500 | 0x44c0000 | +| 84 | blk.9.attn_k.weight | 0xbc5a1500 | 0x226000 | +| 85 | blk.9.attn_norm.weight | 0xbc7c7500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0xbc7cc500 | 0xb40000 | +| 87 | blk.9.attn_q.weight | 0xbd30c500 | 0x898000 | +| 88 | blk.9.attn_v.weight | 0xbdba4500 | 0x2d0000 | +| 89 | blk.9.ffn_down.weight | 0xbde74500 | 0x5a00000 | +| 90 | blk.9.ffn_gate.weight | 0xc3874500 | 0x44c0000 | +| 91 | blk.9.ffn_norm.weight | 0xc7d34500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0xc7d39500 | 0x44c0000 | +| 93 | blk.10.attn_k.weight | 0xcc1f9500 | 0x226000 | +| 94 | blk.10.attn_norm.weight | 0xcc41f500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0xcc424500 | 0xb40000 | +| 96 | blk.10.attn_q.weight | 0xccf64500 | 0x898000 | +| 97 | blk.10.attn_v.weight | 0xcd7fc500 | 0x2d0000 | +| 98 | blk.10.ffn_down.weight | 0xcdacc500 | 0x5a00000 | +| 99 | blk.10.ffn_gate.weight | 0xd34cc500 | 0x44c0000 | +| 100 | blk.10.ffn_norm.weight | 0xd798c500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0xd7991500 | 0x44c0000 | +| 102 | blk.11.attn_k.weight | 0xdbe51500 | 0x226000 | +| 103 | blk.11.attn_norm.weight | 0xdc077500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0xdc07c500 | 0xb40000 | +| 105 | blk.11.attn_q.weight | 0xdcbbc500 | 0x898000 | +| 106 | blk.11.attn_v.weight | 0xdd454500 | 0x2d0000 | +| 107 | blk.11.ffn_down.weight | 0xdd724500 | 0x5a00000 | +| 108 | blk.11.ffn_gate.weight | 0xe3124500 | 0x44c0000 | +| 109 | blk.11.ffn_norm.weight | 0xe75e4500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0xe75e9500 | 0x44c0000 | +| 111 | blk.12.attn_k.weight | 0xebaa9500 | 0x226000 | +| 112 | blk.12.attn_norm.weight | 0xebccf500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0xebcd4500 | 0xb40000 | +| 114 | blk.12.attn_q.weight | 0xec814500 | 0x898000 | +| 115 | blk.12.attn_v.weight | 0xed0ac500 | 0x2d0000 | +| 116 | blk.12.ffn_down.weight | 0xed37c500 | 0x5a00000 | +| 117 | blk.12.ffn_gate.weight | 0xf2d7c500 | 0x44c0000 | +| 118 | blk.12.ffn_norm.weight | 0xf723c500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0xf7241500 | 0x44c0000 | +| 120 | blk.13.attn_k.weight | 0xfb701500 | 0x226000 | +| 121 | blk.13.attn_norm.weight | 0xfb927500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0xfb92c500 | 0xb40000 | +| 123 | blk.13.attn_q.weight | 0xfc46c500 | 0x898000 | +| 124 | blk.13.attn_v.weight | 0xfcd04500 | 0x2d0000 | +| 125 | blk.13.ffn_down.weight | 0xfcfd4500 | 0x5a00000 | +| 126 | blk.13.ffn_gate.weight | 0x1029d4500 | 0x44c0000 | +| 127 | blk.13.ffn_norm.weight | 0x106e94500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0x106e99500 | 0x44c0000 | +| 129 | blk.14.attn_k.weight | 0x10b359500 | 0x226000 | +| 130 | blk.14.attn_norm.weight | 0x10b57f500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0x10b584500 | 0xb40000 | +| 132 | blk.14.attn_q.weight | 0x10c0c4500 | 0x898000 | +| 133 | blk.14.attn_v.weight | 0x10c95c500 | 0x2d0000 | +| 134 | blk.14.ffn_down.weight | 0x10cc2c500 | 0x5a00000 | +| 135 | blk.14.ffn_gate.weight | 0x11262c500 | 0x44c0000 | +| 136 | blk.14.ffn_norm.weight | 0x116aec500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0x116af1500 | 0x44c0000 | +| 138 | blk.15.attn_k.weight | 0x11afb1500 | 0x226000 | +| 139 | blk.15.attn_norm.weight | 0x11b1d7500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0x11b1dc500 | 0xb40000 | +| 141 | blk.15.attn_q.weight | 0x11bd1c500 | 0x898000 | +| 142 | blk.15.attn_v.weight | 0x11c5b4500 | 0x2d0000 | +| 143 | blk.15.ffn_down.weight | 0x11c884500 | 0x5a00000 | +| 144 | blk.15.ffn_gate.weight | 0x122284500 | 0x44c0000 | +| 145 | blk.15.ffn_norm.weight | 0x126744500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0x126749500 | 0x44c0000 | +| 147 | blk.16.attn_k.weight | 0x12ac09500 | 0x226000 | +| 148 | blk.16.attn_norm.weight | 0x12ae2f500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0x12ae34500 | 0xb40000 | +| 150 | blk.16.attn_q.weight | 0x12b974500 | 0x898000 | +| 151 | blk.16.attn_v.weight | 0x12c20c500 | 0x2d0000 | +| 152 | blk.16.ffn_down.weight | 0x12c4dc500 | 0x5a00000 | +| 153 | blk.16.ffn_gate.weight | 0x131edc500 | 0x44c0000 | +| 154 | blk.16.ffn_norm.weight | 0x13639c500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0x1363a1500 | 0x44c0000 | +| 156 | blk.17.attn_k.weight | 0x13a861500 | 0x2d0000 | +| 157 | blk.17.attn_norm.weight | 0x13ab31500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0x13ab36500 | 0xb40000 | +| 159 | blk.17.attn_q.weight | 0x13b676500 | 0xb40000 | +| 160 | blk.17.attn_v.weight | 0x13c1b6500 | 0x2d0000 | +| 161 | blk.17.ffn_down.weight | 0x13c486500 | 0x5a00000 | +| 162 | blk.17.ffn_gate.weight | 0x141e86500 | 0x44c0000 | +| 163 | blk.17.ffn_norm.weight | 0x146346500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0x14634b500 | 0x44c0000 | +| 165 | blk.18.attn_k.weight | 0x14a80b500 | 0x2d0000 | +| 166 | blk.18.attn_norm.weight | 0x14aadb500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0x14aae0500 | 0xb40000 | +| 168 | blk.18.attn_q.weight | 0x14b620500 | 0xb40000 | +| 169 | blk.18.attn_v.weight | 0x14c160500 | 0x2d0000 | +| 170 | blk.18.ffn_down.weight | 0x14c430500 | 0x5a00000 | +| 171 | blk.18.ffn_gate.weight | 0x151e30500 | 0x44c0000 | +| 172 | blk.18.ffn_norm.weight | 0x1562f0500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x1562f5500 | 0x44c0000 | +| 174 | blk.19.attn_k.weight | 0x15a7b5500 | 0x226000 | +| 175 | blk.19.attn_norm.weight | 0x15a9db500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x15a9e0500 | 0xb40000 | +| 177 | blk.19.attn_q.weight | 0x15b520500 | 0x898000 | +| 178 | blk.19.attn_v.weight | 0x15bdb8500 | 0x2d0000 | +| 179 | blk.19.ffn_down.weight | 0x15c088500 | 0x5a00000 | +| 180 | blk.19.ffn_gate.weight | 0x161a88500 | 0x44c0000 | +| 181 | blk.19.ffn_norm.weight | 0x165f48500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x165f4d500 | 0x44c0000 | +| 183 | blk.20.attn_k.weight | 0x16a40d500 | 0x2d0000 | +| 184 | blk.20.attn_norm.weight | 0x16a6dd500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x16a6e2500 | 0xb40000 | +| 186 | blk.20.attn_q.weight | 0x16b222500 | 0xb40000 | +| 187 | blk.20.attn_v.weight | 0x16bd62500 | 0x2d0000 | +| 188 | blk.20.ffn_down.weight | 0x16c032500 | 0x5a00000 | +| 189 | blk.20.ffn_gate.weight | 0x171a32500 | 0x5a00000 | +| 190 | blk.20.ffn_norm.weight | 0x177432500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x177437500 | 0x5a00000 | +| 192 | blk.21.attn_k.weight | 0x17ce37500 | 0x226000 | +| 193 | blk.21.attn_norm.weight | 0x17d05d500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x17d062500 | 0xb40000 | +| 195 | blk.21.attn_q.weight | 0x17dba2500 | 0x898000 | +| 196 | blk.21.attn_v.weight | 0x17e43a500 | 0x2d0000 | +| 197 | blk.21.ffn_down.weight | 0x17e70a500 | 0x5a00000 | +| 198 | blk.21.ffn_gate.weight | 0x18410a500 | 0x5a00000 | +| 199 | blk.21.ffn_norm.weight | 0x189b0a500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x189b0f500 | 0x5a00000 | +| 201 | blk.22.attn_k.weight | 0x18f50f500 | 0x2d0000 | +| 202 | blk.22.attn_norm.weight | 0x18f7df500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x18f7e4500 | 0xb40000 | +| 204 | blk.22.attn_q.weight | 0x190324500 | 0xb40000 | +| 205 | blk.22.attn_v.weight | 0x190e64500 | 0x2d0000 | +| 206 | blk.22.ffn_down.weight | 0x191134500 | 0x5a00000 | +| 207 | blk.22.ffn_gate.weight | 0x196b34500 | 0x5a00000 | +| 208 | blk.22.ffn_norm.weight | 0x19c534500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x19c539500 | 0x5a00000 | +| 210 | blk.23.attn_k.weight | 0x1a1f39500 | 0x2d0000 | +| 211 | blk.23.attn_norm.weight | 0x1a2209500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x1a220e500 | 0xb40000 | +| 213 | blk.23.attn_q.weight | 0x1a2d4e500 | 0xb40000 | +| 214 | blk.23.attn_v.weight | 0x1a388e500 | 0x2d0000 | +| 215 | blk.23.ffn_down.weight | 0x1a3b5e500 | 0x5a00000 | +| 216 | blk.23.ffn_gate.weight | 0x1a955e500 | 0x5a00000 | +| 217 | blk.23.ffn_norm.weight | 0x1aef5e500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x1aef63500 | 0x5a00000 | +| 219 | blk.24.attn_k.weight | 0x1b4963500 | 0x2d0000 | +| 220 | blk.24.attn_norm.weight | 0x1b4c33500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x1b4c38500 | 0xb40000 | +| 222 | blk.24.attn_q.weight | 0x1b5778500 | 0xb40000 | +| 223 | blk.24.attn_v.weight | 0x1b62b8500 | 0x2d0000 | +| 224 | blk.24.ffn_down.weight | 0x1b6588500 | 0x5a00000 | +| 225 | blk.24.ffn_gate.weight | 0x1bbf88500 | 0x5a00000 | +| 226 | blk.24.ffn_norm.weight | 0x1c1988500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x1c198d500 | 0x5a00000 | +| 228 | blk.25.attn_k.weight | 0x1c738d500 | 0x2d0000 | +| 229 | blk.25.attn_norm.weight | 0x1c765d500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x1c7662500 | 0xb40000 | +| 231 | blk.25.attn_q.weight | 0x1c81a2500 | 0xb40000 | +| 232 | blk.25.attn_v.weight | 0x1c8ce2500 | 0x2d0000 | +| 233 | blk.25.ffn_down.weight | 0x1c8fb2500 | 0x5a00000 | +| 234 | blk.25.ffn_gate.weight | 0x1ce9b2500 | 0x5a00000 | +| 235 | blk.25.ffn_norm.weight | 0x1d43b2500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x1d43b7500 | 0x5a00000 | +| 237 | blk.26.attn_k.weight | 0x1d9db7500 | 0x2d0000 | +| 238 | blk.26.attn_norm.weight | 0x1da087500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x1da08c500 | 0xb40000 | +| 240 | blk.26.attn_q.weight | 0x1dabcc500 | 0xb40000 | +| 241 | blk.26.attn_v.weight | 0x1db70c500 | 0x2d0000 | +| 242 | blk.26.ffn_down.weight | 0x1db9dc500 | 0x5a00000 | +| 243 | blk.26.ffn_gate.weight | 0x1e13dc500 | 0x5a00000 | +| 244 | blk.26.ffn_norm.weight | 0x1e6ddc500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x1e6de1500 | 0x5a00000 | +| 246 | blk.27.attn_k.weight | 0x1ec7e1500 | 0x226000 | +| 247 | blk.27.attn_norm.weight | 0x1eca07500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x1eca0c500 | 0xb40000 | +| 249 | blk.27.attn_q.weight | 0x1ed54c500 | 0x898000 | +| 250 | blk.27.attn_v.weight | 0x1edde4500 | 0x2d0000 | +| 251 | blk.27.ffn_down.weight | 0x1ee0b4500 | 0x5a00000 | +| 252 | blk.27.ffn_gate.weight | 0x1f3ab4500 | 0x5a00000 | +| 253 | blk.27.ffn_norm.weight | 0x1f94b4500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x1f94b9500 | 0x5a00000 | +| 255 | blk.28.attn_k.weight | 0x1feeb9500 | 0x2d0000 | +| 256 | blk.28.attn_norm.weight | 0x1ff189500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x1ff18e500 | 0xb40000 | +| 258 | blk.28.attn_q.weight | 0x1ffcce500 | 0xb40000 | +| 259 | blk.28.attn_v.weight | 0x20080e500 | 0x2d0000 | +| 260 | blk.28.ffn_down.weight | 0x200ade500 | 0x5a00000 | +| 261 | blk.28.ffn_gate.weight | 0x2064de500 | 0x5a00000 | +| 262 | blk.28.ffn_norm.weight | 0x20bede500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x20bee3500 | 0x5a00000 | +| 264 | blk.29.attn_k.weight | 0x2118e3500 | 0x2d0000 | +| 265 | blk.29.attn_norm.weight | 0x211bb3500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x211bb8500 | 0xb40000 | +| 267 | blk.29.attn_q.weight | 0x2126f8500 | 0xb40000 | +| 268 | blk.29.attn_v.weight | 0x213238500 | 0x2d0000 | +| 269 | blk.29.ffn_down.weight | 0x213508500 | 0x5a00000 | +| 270 | blk.29.ffn_gate.weight | 0x218f08500 | 0x5a00000 | +| 271 | blk.29.ffn_norm.weight | 0x21e908500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x21e90d500 | 0x5a00000 | +| 273 | blk.30.attn_k.weight | 0x22430d500 | 0x2d0000 | +| 274 | blk.30.attn_norm.weight | 0x2245dd500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x2245e2500 | 0xb40000 | +| 276 | blk.30.attn_q.weight | 0x225122500 | 0xb40000 | +| 277 | blk.30.attn_v.weight | 0x225c62500 | 0x2d0000 | +| 278 | blk.30.ffn_down.weight | 0x225f32500 | 0x5a00000 | +| 279 | blk.30.ffn_gate.weight | 0x22b932500 | 0x5a00000 | +| 280 | blk.30.ffn_norm.weight | 0x231332500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x231337500 | 0x5a00000 | +| 282 | blk.31.attn_k.weight | 0x236d37500 | 0x2d0000 | +| 283 | blk.31.attn_norm.weight | 0x237007500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x23700c500 | 0xb40000 | +| 285 | blk.31.attn_q.weight | 0x237b4c500 | 0xb40000 | +| 286 | blk.31.attn_v.weight | 0x23868c500 | 0x2d0000 | +| 287 | blk.31.ffn_down.weight | 0x23895c500 | 0x5a00000 | +| 288 | blk.31.ffn_gate.weight | 0x23e35c500 | 0x5a00000 | +| 289 | blk.31.ffn_norm.weight | 0x243d5c500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x243d61500 | 0x5a00000 | +| 291 | blk.32.attn_k.weight | 0x249761500 | 0x2d0000 | +| 292 | blk.32.attn_norm.weight | 0x249a31500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x249a36500 | 0xb40000 | +| 294 | blk.32.attn_q.weight | 0x24a576500 | 0xb40000 | +| 295 | blk.32.attn_v.weight | 0x24b0b6500 | 0x2d0000 | +| 296 | blk.32.ffn_down.weight | 0x24b386500 | 0x5a00000 | +| 297 | blk.32.ffn_gate.weight | 0x250d86500 | 0x5a00000 | +| 298 | blk.32.ffn_norm.weight | 0x256786500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x25678b500 | 0x5a00000 | +| 300 | blk.33.attn_k.weight | 0x25c18b500 | 0x2d0000 | +| 301 | blk.33.attn_norm.weight | 0x25c45b500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x25c460500 | 0xb40000 | +| 303 | blk.33.attn_q.weight | 0x25cfa0500 | 0xb40000 | +| 304 | blk.33.attn_v.weight | 0x25dae0500 | 0x2d0000 | +| 305 | blk.33.ffn_down.weight | 0x25ddb0500 | 0x5a00000 | +| 306 | blk.33.ffn_gate.weight | 0x2637b0500 | 0x5a00000 | +| 307 | blk.33.ffn_norm.weight | 0x2691b0500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x2691b5500 | 0x5a00000 | +| 309 | blk.34.attn_k.weight | 0x26ebb5500 | 0x2d0000 | +| 310 | blk.34.attn_norm.weight | 0x26ee85500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x26ee8a500 | 0xb40000 | +| 312 | blk.34.attn_q.weight | 0x26f9ca500 | 0xb40000 | +| 313 | blk.34.attn_v.weight | 0x27050a500 | 0x2d0000 | +| 314 | blk.34.ffn_down.weight | 0x2707da500 | 0x5a00000 | +| 315 | blk.34.ffn_gate.weight | 0x2761da500 | 0x5a00000 | +| 316 | blk.34.ffn_norm.weight | 0x27bbda500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x27bbdf500 | 0x5a00000 | +| 318 | blk.35.attn_k.weight | 0x2815df500 | 0x2d0000 | +| 319 | blk.35.attn_norm.weight | 0x2818af500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x2818b4500 | 0xb40000 | +| 321 | blk.35.attn_q.weight | 0x2823f4500 | 0xb40000 | +| 322 | blk.35.attn_v.weight | 0x282f34500 | 0x2d0000 | +| 323 | blk.35.ffn_down.weight | 0x283204500 | 0x5a00000 | +| 324 | blk.35.ffn_gate.weight | 0x288c04500 | 0x5a00000 | +| 325 | blk.35.ffn_norm.weight | 0x28e604500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x28e609500 | 0x5a00000 | +| 327 | blk.36.attn_k.weight | 0x294009500 | 0x2d0000 | +| 328 | blk.36.attn_norm.weight | 0x2942d9500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x2942de500 | 0xb40000 | +| 330 | blk.36.attn_q.weight | 0x294e1e500 | 0xb40000 | +| 331 | blk.36.attn_v.weight | 0x29595e500 | 0x2d0000 | +| 332 | blk.36.ffn_down.weight | 0x295c2e500 | 0x5a00000 | +| 333 | blk.36.ffn_gate.weight | 0x29b62e500 | 0x5a00000 | +| 334 | blk.36.ffn_norm.weight | 0x2a102e500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x2a1033500 | 0x5a00000 | +| 336 | blk.37.attn_k.weight | 0x2a6a33500 | 0x2d0000 | +| 337 | blk.37.attn_norm.weight | 0x2a6d03500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x2a6d08500 | 0xb40000 | +| 339 | blk.37.attn_q.weight | 0x2a7848500 | 0xb40000 | +| 340 | blk.37.attn_v.weight | 0x2a8388500 | 0x2d0000 | +| 341 | blk.37.ffn_down.weight | 0x2a8658500 | 0x5a00000 | +| 342 | blk.37.ffn_gate.weight | 0x2ae058500 | 0x5a00000 | +| 343 | blk.37.ffn_norm.weight | 0x2b3a58500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x2b3a5d500 | 0x5a00000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q4_K | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-Q5_K_M.md b/scores/Dolphin-Mistral-24B-Venice-Edition-Q5_K_M.md new file mode 100644 index 0000000..eafa3ec --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-Q5_K_M.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-Q5_K_M.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 17 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-Q5\_K\_M.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-q5_k_mgguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0x1b800000 | +| 1 | output_norm.weight | 0x1bf84500 | 0x5000 | +| 2 | token_embd.weight | 0x1bf89500 | 0x11300000 | +| 3 | blk.0.attn_k.weight | 0x2d289500 | 0x2d0000 | +| 4 | blk.0.attn_norm.weight | 0x2d559500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x2d55e500 | 0xdc0000 | +| 6 | blk.0.attn_q.weight | 0x2e31e500 | 0xb40000 | +| 7 | blk.0.attn_v.weight | 0x2ee5e500 | 0x370000 | +| 8 | blk.0.ffn_down.weight | 0x2f1ce500 | 0x8340000 | +| 9 | blk.0.ffn_gate.weight | 0x3750e500 | 0x5a00000 | +| 10 | blk.0.ffn_norm.weight | 0x3cf0e500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x3cf13500 | 0x5a00000 | +| 12 | blk.1.attn_k.weight | 0x42913500 | 0x2d0000 | +| 13 | blk.1.attn_norm.weight | 0x42be3500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x42be8500 | 0xdc0000 | +| 15 | blk.1.attn_q.weight | 0x439a8500 | 0xb40000 | +| 16 | blk.1.attn_v.weight | 0x444e8500 | 0x370000 | +| 17 | blk.1.ffn_down.weight | 0x44858500 | 0x8340000 | +| 18 | blk.1.ffn_gate.weight | 0x4cb98500 | 0x5a00000 | +| 19 | blk.1.ffn_norm.weight | 0x52598500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x5259d500 | 0x5a00000 | +| 21 | blk.2.attn_k.weight | 0x57f9d500 | 0x2d0000 | +| 22 | blk.2.attn_norm.weight | 0x5826d500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x58272500 | 0xdc0000 | +| 24 | blk.2.attn_q.weight | 0x59032500 | 0xb40000 | +| 25 | blk.2.attn_v.weight | 0x59b72500 | 0x370000 | +| 26 | blk.2.ffn_down.weight | 0x59ee2500 | 0x8340000 | +| 27 | blk.2.ffn_gate.weight | 0x62222500 | 0x5a00000 | +| 28 | blk.2.ffn_norm.weight | 0x67c22500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x67c27500 | 0x5a00000 | +| 30 | blk.3.attn_k.weight | 0x6d627500 | 0x2d0000 | +| 31 | blk.3.attn_norm.weight | 0x6d8f7500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x6d8fc500 | 0xdc0000 | +| 33 | blk.3.attn_q.weight | 0x6e6bc500 | 0xb40000 | +| 34 | blk.3.attn_v.weight | 0x6f1fc500 | 0x370000 | +| 35 | blk.3.ffn_down.weight | 0x6f56c500 | 0x8340000 | +| 36 | blk.3.ffn_gate.weight | 0x778ac500 | 0x5a00000 | +| 37 | blk.3.ffn_norm.weight | 0x7d2ac500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0x7d2b1500 | 0x5a00000 | +| 39 | blk.4.attn_k.weight | 0x82cb1500 | 0x2d0000 | +| 40 | blk.4.attn_norm.weight | 0x82f81500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0x82f86500 | 0xdc0000 | +| 42 | blk.4.attn_q.weight | 0x83d46500 | 0xb40000 | +| 43 | blk.4.attn_v.weight | 0x84886500 | 0x41a000 | +| 44 | blk.4.ffn_down.weight | 0x84ca0500 | 0x8340000 | +| 45 | blk.4.ffn_gate.weight | 0x8cfe0500 | 0x5a00000 | +| 46 | blk.4.ffn_norm.weight | 0x929e0500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0x929e5500 | 0x5a00000 | +| 48 | blk.5.attn_k.weight | 0x983e5500 | 0x2d0000 | +| 49 | blk.5.attn_norm.weight | 0x986b5500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0x986ba500 | 0xdc0000 | +| 51 | blk.5.attn_q.weight | 0x9947a500 | 0xb40000 | +| 52 | blk.5.attn_v.weight | 0x99fba500 | 0x41a000 | +| 53 | blk.5.ffn_down.weight | 0x9a3d4500 | 0x6e00000 | +| 54 | blk.5.ffn_gate.weight | 0xa11d4500 | 0x5a00000 | +| 55 | blk.5.ffn_norm.weight | 0xa6bd4500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0xa6bd9500 | 0x5a00000 | +| 57 | blk.6.attn_k.weight | 0xac5d9500 | 0x2d0000 | +| 58 | blk.6.attn_norm.weight | 0xac8a9500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0xac8ae500 | 0xdc0000 | +| 60 | blk.6.attn_q.weight | 0xad66e500 | 0xb40000 | +| 61 | blk.6.attn_v.weight | 0xae1ae500 | 0x370000 | +| 62 | blk.6.ffn_down.weight | 0xae51e500 | 0x6e00000 | +| 63 | blk.6.ffn_gate.weight | 0xb531e500 | 0x5a00000 | +| 64 | blk.6.ffn_norm.weight | 0xbad1e500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0xbad23500 | 0x5a00000 | +| 66 | blk.7.attn_k.weight | 0xc0723500 | 0x2d0000 | +| 67 | blk.7.attn_norm.weight | 0xc09f3500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0xc09f8500 | 0xdc0000 | +| 69 | blk.7.attn_q.weight | 0xc17b8500 | 0xb40000 | +| 70 | blk.7.attn_v.weight | 0xc22f8500 | 0x41a000 | +| 71 | blk.7.ffn_down.weight | 0xc2712500 | 0x8340000 | +| 72 | blk.7.ffn_gate.weight | 0xcaa52500 | 0x5a00000 | +| 73 | blk.7.ffn_norm.weight | 0xd0452500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0xd0457500 | 0x5a00000 | +| 75 | blk.8.attn_k.weight | 0xd5e57500 | 0x2d0000 | +| 76 | blk.8.attn_norm.weight | 0xd6127500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0xd612c500 | 0xdc0000 | +| 78 | blk.8.attn_q.weight | 0xd6eec500 | 0xb40000 | +| 79 | blk.8.attn_v.weight | 0xd7a2c500 | 0x41a000 | +| 80 | blk.8.ffn_down.weight | 0xd7e46500 | 0x6e00000 | +| 81 | blk.8.ffn_gate.weight | 0xdec46500 | 0x5a00000 | +| 82 | blk.8.ffn_norm.weight | 0xe4646500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0xe464b500 | 0x5a00000 | +| 84 | blk.9.attn_k.weight | 0xea04b500 | 0x2d0000 | +| 85 | blk.9.attn_norm.weight | 0xea31b500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0xea320500 | 0xdc0000 | +| 87 | blk.9.attn_q.weight | 0xeb0e0500 | 0xb40000 | +| 88 | blk.9.attn_v.weight | 0xebc20500 | 0x370000 | +| 89 | blk.9.ffn_down.weight | 0xebf90500 | 0x6e00000 | +| 90 | blk.9.ffn_gate.weight | 0xf2d90500 | 0x5a00000 | +| 91 | blk.9.ffn_norm.weight | 0xf8790500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0xf8795500 | 0x5a00000 | +| 93 | blk.10.attn_k.weight | 0xfe195500 | 0x2d0000 | +| 94 | blk.10.attn_norm.weight | 0xfe465500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0xfe46a500 | 0xdc0000 | +| 96 | blk.10.attn_q.weight | 0xff22a500 | 0xb40000 | +| 97 | blk.10.attn_v.weight | 0xffd6a500 | 0x41a000 | +| 98 | blk.10.ffn_down.weight | 0x100184500 | 0x8340000 | +| 99 | blk.10.ffn_gate.weight | 0x1084c4500 | 0x5a00000 | +| 100 | blk.10.ffn_norm.weight | 0x10dec4500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0x10dec9500 | 0x5a00000 | +| 102 | blk.11.attn_k.weight | 0x1138c9500 | 0x2d0000 | +| 103 | blk.11.attn_norm.weight | 0x113b99500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0x113b9e500 | 0xdc0000 | +| 105 | blk.11.attn_q.weight | 0x11495e500 | 0xb40000 | +| 106 | blk.11.attn_v.weight | 0x11549e500 | 0x41a000 | +| 107 | blk.11.ffn_down.weight | 0x1158b8500 | 0x6e00000 | +| 108 | blk.11.ffn_gate.weight | 0x11c6b8500 | 0x5a00000 | +| 109 | blk.11.ffn_norm.weight | 0x1220b8500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0x1220bd500 | 0x5a00000 | +| 111 | blk.12.attn_k.weight | 0x127abd500 | 0x2d0000 | +| 112 | blk.12.attn_norm.weight | 0x127d8d500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0x127d92500 | 0xdc0000 | +| 114 | blk.12.attn_q.weight | 0x128b52500 | 0xb40000 | +| 115 | blk.12.attn_v.weight | 0x129692500 | 0x370000 | +| 116 | blk.12.ffn_down.weight | 0x129a02500 | 0x6e00000 | +| 117 | blk.12.ffn_gate.weight | 0x130802500 | 0x5a00000 | +| 118 | blk.12.ffn_norm.weight | 0x136202500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0x136207500 | 0x5a00000 | +| 120 | blk.13.attn_k.weight | 0x13bc07500 | 0x2d0000 | +| 121 | blk.13.attn_norm.weight | 0x13bed7500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0x13bedc500 | 0xdc0000 | +| 123 | blk.13.attn_q.weight | 0x13cc9c500 | 0xb40000 | +| 124 | blk.13.attn_v.weight | 0x13d7dc500 | 0x41a000 | +| 125 | blk.13.ffn_down.weight | 0x13dbf6500 | 0x8340000 | +| 126 | blk.13.ffn_gate.weight | 0x145f36500 | 0x5a00000 | +| 127 | blk.13.ffn_norm.weight | 0x14b936500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0x14b93b500 | 0x5a00000 | +| 129 | blk.14.attn_k.weight | 0x15133b500 | 0x2d0000 | +| 130 | blk.14.attn_norm.weight | 0x15160b500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0x151610500 | 0xdc0000 | +| 132 | blk.14.attn_q.weight | 0x1523d0500 | 0xb40000 | +| 133 | blk.14.attn_v.weight | 0x152f10500 | 0x41a000 | +| 134 | blk.14.ffn_down.weight | 0x15332a500 | 0x6e00000 | +| 135 | blk.14.ffn_gate.weight | 0x15a12a500 | 0x5a00000 | +| 136 | blk.14.ffn_norm.weight | 0x15fb2a500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0x15fb2f500 | 0x5a00000 | +| 138 | blk.15.attn_k.weight | 0x16552f500 | 0x2d0000 | +| 139 | blk.15.attn_norm.weight | 0x1657ff500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0x165804500 | 0xdc0000 | +| 141 | blk.15.attn_q.weight | 0x1665c4500 | 0xb40000 | +| 142 | blk.15.attn_v.weight | 0x167104500 | 0x370000 | +| 143 | blk.15.ffn_down.weight | 0x167474500 | 0x6e00000 | +| 144 | blk.15.ffn_gate.weight | 0x16e274500 | 0x5a00000 | +| 145 | blk.15.ffn_norm.weight | 0x173c74500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0x173c79500 | 0x5a00000 | +| 147 | blk.16.attn_k.weight | 0x179679500 | 0x2d0000 | +| 148 | blk.16.attn_norm.weight | 0x179949500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0x17994e500 | 0xdc0000 | +| 150 | blk.16.attn_q.weight | 0x17a70e500 | 0xb40000 | +| 151 | blk.16.attn_v.weight | 0x17b24e500 | 0x41a000 | +| 152 | blk.16.ffn_down.weight | 0x17b668500 | 0x8340000 | +| 153 | blk.16.ffn_gate.weight | 0x1839a8500 | 0x5a00000 | +| 154 | blk.16.ffn_norm.weight | 0x1893a8500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0x1893ad500 | 0x5a00000 | +| 156 | blk.17.attn_k.weight | 0x18edad500 | 0x2d0000 | +| 157 | blk.17.attn_norm.weight | 0x18f07d500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0x18f082500 | 0xdc0000 | +| 159 | blk.17.attn_q.weight | 0x18fe42500 | 0xb40000 | +| 160 | blk.17.attn_v.weight | 0x190982500 | 0x41a000 | +| 161 | blk.17.ffn_down.weight | 0x190d9c500 | 0x6e00000 | +| 162 | blk.17.ffn_gate.weight | 0x197b9c500 | 0x5a00000 | +| 163 | blk.17.ffn_norm.weight | 0x19d59c500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0x19d5a1500 | 0x5a00000 | +| 165 | blk.18.attn_k.weight | 0x1a2fa1500 | 0x2d0000 | +| 166 | blk.18.attn_norm.weight | 0x1a3271500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0x1a3276500 | 0xdc0000 | +| 168 | blk.18.attn_q.weight | 0x1a4036500 | 0xb40000 | +| 169 | blk.18.attn_v.weight | 0x1a4b76500 | 0x370000 | +| 170 | blk.18.ffn_down.weight | 0x1a4ee6500 | 0x6e00000 | +| 171 | blk.18.ffn_gate.weight | 0x1abce6500 | 0x5a00000 | +| 172 | blk.18.ffn_norm.weight | 0x1b16e6500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x1b16eb500 | 0x5a00000 | +| 174 | blk.19.attn_k.weight | 0x1b70eb500 | 0x2d0000 | +| 175 | blk.19.attn_norm.weight | 0x1b73bb500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x1b73c0500 | 0xdc0000 | +| 177 | blk.19.attn_q.weight | 0x1b8180500 | 0xb40000 | +| 178 | blk.19.attn_v.weight | 0x1b8cc0500 | 0x41a000 | +| 179 | blk.19.ffn_down.weight | 0x1b90da500 | 0x8340000 | +| 180 | blk.19.ffn_gate.weight | 0x1c141a500 | 0x5a00000 | +| 181 | blk.19.ffn_norm.weight | 0x1c6e1a500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x1c6e1f500 | 0x5a00000 | +| 183 | blk.20.attn_k.weight | 0x1cc81f500 | 0x2d0000 | +| 184 | blk.20.attn_norm.weight | 0x1ccaef500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x1ccaf4500 | 0xdc0000 | +| 186 | blk.20.attn_q.weight | 0x1cd8b4500 | 0xb40000 | +| 187 | blk.20.attn_v.weight | 0x1ce3f4500 | 0x41a000 | +| 188 | blk.20.ffn_down.weight | 0x1ce80e500 | 0x6e00000 | +| 189 | blk.20.ffn_gate.weight | 0x1d560e500 | 0x5a00000 | +| 190 | blk.20.ffn_norm.weight | 0x1db00e500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x1db013500 | 0x5a00000 | +| 192 | blk.21.attn_k.weight | 0x1e0a13500 | 0x2d0000 | +| 193 | blk.21.attn_norm.weight | 0x1e0ce3500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x1e0ce8500 | 0xdc0000 | +| 195 | blk.21.attn_q.weight | 0x1e1aa8500 | 0xb40000 | +| 196 | blk.21.attn_v.weight | 0x1e25e8500 | 0x370000 | +| 197 | blk.21.ffn_down.weight | 0x1e2958500 | 0x6e00000 | +| 198 | blk.21.ffn_gate.weight | 0x1e9758500 | 0x5a00000 | +| 199 | blk.21.ffn_norm.weight | 0x1ef158500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x1ef15d500 | 0x5a00000 | +| 201 | blk.22.attn_k.weight | 0x1f4b5d500 | 0x2d0000 | +| 202 | blk.22.attn_norm.weight | 0x1f4e2d500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x1f4e32500 | 0xdc0000 | +| 204 | blk.22.attn_q.weight | 0x1f5bf2500 | 0xb40000 | +| 205 | blk.22.attn_v.weight | 0x1f6732500 | 0x41a000 | +| 206 | blk.22.ffn_down.weight | 0x1f6b4c500 | 0x8340000 | +| 207 | blk.22.ffn_gate.weight | 0x1fee8c500 | 0x5a00000 | +| 208 | blk.22.ffn_norm.weight | 0x20488c500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x204891500 | 0x5a00000 | +| 210 | blk.23.attn_k.weight | 0x20a291500 | 0x2d0000 | +| 211 | blk.23.attn_norm.weight | 0x20a561500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x20a566500 | 0xdc0000 | +| 213 | blk.23.attn_q.weight | 0x20b326500 | 0xb40000 | +| 214 | blk.23.attn_v.weight | 0x20be66500 | 0x41a000 | +| 215 | blk.23.ffn_down.weight | 0x20c280500 | 0x6e00000 | +| 216 | blk.23.ffn_gate.weight | 0x213080500 | 0x5a00000 | +| 217 | blk.23.ffn_norm.weight | 0x218a80500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x218a85500 | 0x5a00000 | +| 219 | blk.24.attn_k.weight | 0x21e485500 | 0x2d0000 | +| 220 | blk.24.attn_norm.weight | 0x21e755500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x21e75a500 | 0xdc0000 | +| 222 | blk.24.attn_q.weight | 0x21f51a500 | 0xb40000 | +| 223 | blk.24.attn_v.weight | 0x22005a500 | 0x370000 | +| 224 | blk.24.ffn_down.weight | 0x2203ca500 | 0x6e00000 | +| 225 | blk.24.ffn_gate.weight | 0x2271ca500 | 0x5a00000 | +| 226 | blk.24.ffn_norm.weight | 0x22cbca500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x22cbcf500 | 0x5a00000 | +| 228 | blk.25.attn_k.weight | 0x2325cf500 | 0x370000 | +| 229 | blk.25.attn_norm.weight | 0x23293f500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x232944500 | 0xdc0000 | +| 231 | blk.25.attn_q.weight | 0x233704500 | 0xdc0000 | +| 232 | blk.25.attn_v.weight | 0x2344c4500 | 0x41a000 | +| 233 | blk.25.ffn_down.weight | 0x2348de500 | 0x8340000 | +| 234 | blk.25.ffn_gate.weight | 0x23cc1e500 | 0x5a00000 | +| 235 | blk.25.ffn_norm.weight | 0x24261e500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x242623500 | 0x5a00000 | +| 237 | blk.26.attn_k.weight | 0x248023500 | 0x2d0000 | +| 238 | blk.26.attn_norm.weight | 0x2482f3500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x2482f8500 | 0xdc0000 | +| 240 | blk.26.attn_q.weight | 0x2490b8500 | 0xb40000 | +| 241 | blk.26.attn_v.weight | 0x249bf8500 | 0x41a000 | +| 242 | blk.26.ffn_down.weight | 0x24a012500 | 0x6e00000 | +| 243 | blk.26.ffn_gate.weight | 0x250e12500 | 0x5a00000 | +| 244 | blk.26.ffn_norm.weight | 0x256812500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x256817500 | 0x5a00000 | +| 246 | blk.27.attn_k.weight | 0x25c217500 | 0x2d0000 | +| 247 | blk.27.attn_norm.weight | 0x25c4e7500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x25c4ec500 | 0xdc0000 | +| 249 | blk.27.attn_q.weight | 0x25d2ac500 | 0xb40000 | +| 250 | blk.27.attn_v.weight | 0x25ddec500 | 0x370000 | +| 251 | blk.27.ffn_down.weight | 0x25e15c500 | 0x6e00000 | +| 252 | blk.27.ffn_gate.weight | 0x264f5c500 | 0x5a00000 | +| 253 | blk.27.ffn_norm.weight | 0x26a95c500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x26a961500 | 0x5a00000 | +| 255 | blk.28.attn_k.weight | 0x270361500 | 0x370000 | +| 256 | blk.28.attn_norm.weight | 0x2706d1500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x2706d6500 | 0xdc0000 | +| 258 | blk.28.attn_q.weight | 0x271496500 | 0xdc0000 | +| 259 | blk.28.attn_v.weight | 0x272256500 | 0x41a000 | +| 260 | blk.28.ffn_down.weight | 0x272670500 | 0x8340000 | +| 261 | blk.28.ffn_gate.weight | 0x27a9b0500 | 0x5a00000 | +| 262 | blk.28.ffn_norm.weight | 0x2803b0500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x2803b5500 | 0x5a00000 | +| 264 | blk.29.attn_k.weight | 0x285db5500 | 0x2d0000 | +| 265 | blk.29.attn_norm.weight | 0x286085500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x28608a500 | 0xdc0000 | +| 267 | blk.29.attn_q.weight | 0x286e4a500 | 0xb40000 | +| 268 | blk.29.attn_v.weight | 0x28798a500 | 0x41a000 | +| 269 | blk.29.ffn_down.weight | 0x287da4500 | 0x6e00000 | +| 270 | blk.29.ffn_gate.weight | 0x28eba4500 | 0x5a00000 | +| 271 | blk.29.ffn_norm.weight | 0x2945a4500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x2945a9500 | 0x5a00000 | +| 273 | blk.30.attn_k.weight | 0x299fa9500 | 0x370000 | +| 274 | blk.30.attn_norm.weight | 0x29a319500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x29a31e500 | 0xdc0000 | +| 276 | blk.30.attn_q.weight | 0x29b0de500 | 0xdc0000 | +| 277 | blk.30.attn_v.weight | 0x29be9e500 | 0x41a000 | +| 278 | blk.30.ffn_down.weight | 0x29c2b8500 | 0x6e00000 | +| 279 | blk.30.ffn_gate.weight | 0x2a30b8500 | 0x6e00000 | +| 280 | blk.30.ffn_norm.weight | 0x2a9eb8500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x2a9ebd500 | 0x6e00000 | +| 282 | blk.31.attn_k.weight | 0x2b0cbd500 | 0x2d0000 | +| 283 | blk.31.attn_norm.weight | 0x2b0f8d500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x2b0f92500 | 0xdc0000 | +| 285 | blk.31.attn_q.weight | 0x2b1d52500 | 0xb40000 | +| 286 | blk.31.attn_v.weight | 0x2b2892500 | 0x41a000 | +| 287 | blk.31.ffn_down.weight | 0x2b2cac500 | 0x8340000 | +| 288 | blk.31.ffn_gate.weight | 0x2bafec500 | 0x6e00000 | +| 289 | blk.31.ffn_norm.weight | 0x2c1dec500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x2c1df1500 | 0x6e00000 | +| 291 | blk.32.attn_k.weight | 0x2c8bf1500 | 0x370000 | +| 292 | blk.32.attn_norm.weight | 0x2c8f61500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x2c8f66500 | 0xdc0000 | +| 294 | blk.32.attn_q.weight | 0x2c9d26500 | 0xdc0000 | +| 295 | blk.32.attn_v.weight | 0x2caae6500 | 0x41a000 | +| 296 | blk.32.ffn_down.weight | 0x2caf00500 | 0x6e00000 | +| 297 | blk.32.ffn_gate.weight | 0x2d1d00500 | 0x6e00000 | +| 298 | blk.32.ffn_norm.weight | 0x2d8b00500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x2d8b05500 | 0x6e00000 | +| 300 | blk.33.attn_k.weight | 0x2df905500 | 0x370000 | +| 301 | blk.33.attn_norm.weight | 0x2dfc75500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x2dfc7a500 | 0xdc0000 | +| 303 | blk.33.attn_q.weight | 0x2e0a3a500 | 0xdc0000 | +| 304 | blk.33.attn_v.weight | 0x2e17fa500 | 0x41a000 | +| 305 | blk.33.ffn_down.weight | 0x2e1c14500 | 0x6e00000 | +| 306 | blk.33.ffn_gate.weight | 0x2e8a14500 | 0x6e00000 | +| 307 | blk.33.ffn_norm.weight | 0x2ef814500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x2ef819500 | 0x6e00000 | +| 309 | blk.34.attn_k.weight | 0x2f6619500 | 0x370000 | +| 310 | blk.34.attn_norm.weight | 0x2f6989500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x2f698e500 | 0xdc0000 | +| 312 | blk.34.attn_q.weight | 0x2f774e500 | 0xdc0000 | +| 313 | blk.34.attn_v.weight | 0x2f850e500 | 0x41a000 | +| 314 | blk.34.ffn_down.weight | 0x2f8928500 | 0x8340000 | +| 315 | blk.34.ffn_gate.weight | 0x300c68500 | 0x6e00000 | +| 316 | blk.34.ffn_norm.weight | 0x307a68500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x307a6d500 | 0x6e00000 | +| 318 | blk.35.attn_k.weight | 0x30e86d500 | 0x370000 | +| 319 | blk.35.attn_norm.weight | 0x30ebdd500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x30ebe2500 | 0xdc0000 | +| 321 | blk.35.attn_q.weight | 0x30f9a2500 | 0xdc0000 | +| 322 | blk.35.attn_v.weight | 0x310762500 | 0x41a000 | +| 323 | blk.35.ffn_down.weight | 0x310b7c500 | 0x8340000 | +| 324 | blk.35.ffn_gate.weight | 0x318ebc500 | 0x6e00000 | +| 325 | blk.35.ffn_norm.weight | 0x31fcbc500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x31fcc1500 | 0x6e00000 | +| 327 | blk.36.attn_k.weight | 0x326ac1500 | 0x370000 | +| 328 | blk.36.attn_norm.weight | 0x326e31500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x326e36500 | 0xdc0000 | +| 330 | blk.36.attn_q.weight | 0x327bf6500 | 0xdc0000 | +| 331 | blk.36.attn_v.weight | 0x3289b6500 | 0x41a000 | +| 332 | blk.36.ffn_down.weight | 0x328dd0500 | 0x8340000 | +| 333 | blk.36.ffn_gate.weight | 0x331110500 | 0x6e00000 | +| 334 | blk.36.ffn_norm.weight | 0x337f10500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x337f15500 | 0x6e00000 | +| 336 | blk.37.attn_k.weight | 0x33ed15500 | 0x370000 | +| 337 | blk.37.attn_norm.weight | 0x33f085500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x33f08a500 | 0xdc0000 | +| 339 | blk.37.attn_q.weight | 0x33fe4a500 | 0xdc0000 | +| 340 | blk.37.attn_v.weight | 0x340c0a500 | 0x41a000 | +| 341 | blk.37.ffn_down.weight | 0x341024500 | 0x8340000 | +| 342 | blk.37.ffn_gate.weight | 0x349364500 | 0x6e00000 | +| 343 | blk.37.ffn_norm.weight | 0x350164500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x350169500 | 0x6e00000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q5_K | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-Q5_K_S.md b/scores/Dolphin-Mistral-24B-Venice-Edition-Q5_K_S.md new file mode 100644 index 0000000..6c99376 --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-Q5_K_S.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-Q5_K_S.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 16 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-Q5\_K\_S.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-q5_k_sgguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0x1b800000 | +| 1 | output_norm.weight | 0x1bf84500 | 0x5000 | +| 2 | token_embd.weight | 0x1bf89500 | 0x11300000 | +| 3 | blk.0.attn_k.weight | 0x2d289500 | 0x2d0000 | +| 4 | blk.0.attn_norm.weight | 0x2d559500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x2d55e500 | 0xdc0000 | +| 6 | blk.0.attn_q.weight | 0x2e31e500 | 0xb40000 | +| 7 | blk.0.attn_v.weight | 0x2ee5e500 | 0x2d0000 | +| 8 | blk.0.ffn_down.weight | 0x2f12e500 | 0x6e00000 | +| 9 | blk.0.ffn_gate.weight | 0x35f2e500 | 0x5a00000 | +| 10 | blk.0.ffn_norm.weight | 0x3b92e500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x3b933500 | 0x5a00000 | +| 12 | blk.1.attn_k.weight | 0x41333500 | 0x2d0000 | +| 13 | blk.1.attn_norm.weight | 0x41603500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x41608500 | 0xdc0000 | +| 15 | blk.1.attn_q.weight | 0x423c8500 | 0xb40000 | +| 16 | blk.1.attn_v.weight | 0x42f08500 | 0x2d0000 | +| 17 | blk.1.ffn_down.weight | 0x431d8500 | 0x6e00000 | +| 18 | blk.1.ffn_gate.weight | 0x49fd8500 | 0x5a00000 | +| 19 | blk.1.ffn_norm.weight | 0x4f9d8500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x4f9dd500 | 0x5a00000 | +| 21 | blk.2.attn_k.weight | 0x553dd500 | 0x2d0000 | +| 22 | blk.2.attn_norm.weight | 0x556ad500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x556b2500 | 0xdc0000 | +| 24 | blk.2.attn_q.weight | 0x56472500 | 0xb40000 | +| 25 | blk.2.attn_v.weight | 0x56fb2500 | 0x2d0000 | +| 26 | blk.2.ffn_down.weight | 0x57282500 | 0x6e00000 | +| 27 | blk.2.ffn_gate.weight | 0x5e082500 | 0x5a00000 | +| 28 | blk.2.ffn_norm.weight | 0x63a82500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x63a87500 | 0x5a00000 | +| 30 | blk.3.attn_k.weight | 0x69487500 | 0x2d0000 | +| 31 | blk.3.attn_norm.weight | 0x69757500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x6975c500 | 0xdc0000 | +| 33 | blk.3.attn_q.weight | 0x6a51c500 | 0xb40000 | +| 34 | blk.3.attn_v.weight | 0x6b05c500 | 0x2d0000 | +| 35 | blk.3.ffn_down.weight | 0x6b32c500 | 0x6e00000 | +| 36 | blk.3.ffn_gate.weight | 0x7212c500 | 0x5a00000 | +| 37 | blk.3.ffn_norm.weight | 0x77b2c500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0x77b31500 | 0x5a00000 | +| 39 | blk.4.attn_k.weight | 0x7d531500 | 0x2d0000 | +| 40 | blk.4.attn_norm.weight | 0x7d801500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0x7d806500 | 0xdc0000 | +| 42 | blk.4.attn_q.weight | 0x7e5c6500 | 0xb40000 | +| 43 | blk.4.attn_v.weight | 0x7f106500 | 0x2d0000 | +| 44 | blk.4.ffn_down.weight | 0x7f3d6500 | 0x6e00000 | +| 45 | blk.4.ffn_gate.weight | 0x861d6500 | 0x5a00000 | +| 46 | blk.4.ffn_norm.weight | 0x8bbd6500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0x8bbdb500 | 0x5a00000 | +| 48 | blk.5.attn_k.weight | 0x915db500 | 0x2d0000 | +| 49 | blk.5.attn_norm.weight | 0x918ab500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0x918b0500 | 0xdc0000 | +| 51 | blk.5.attn_q.weight | 0x92670500 | 0xb40000 | +| 52 | blk.5.attn_v.weight | 0x931b0500 | 0x2d0000 | +| 53 | blk.5.ffn_down.weight | 0x93480500 | 0x6e00000 | +| 54 | blk.5.ffn_gate.weight | 0x9a280500 | 0x5a00000 | +| 55 | blk.5.ffn_norm.weight | 0x9fc80500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0x9fc85500 | 0x5a00000 | +| 57 | blk.6.attn_k.weight | 0xa5685500 | 0x2d0000 | +| 58 | blk.6.attn_norm.weight | 0xa5955500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0xa595a500 | 0xdc0000 | +| 60 | blk.6.attn_q.weight | 0xa671a500 | 0xb40000 | +| 61 | blk.6.attn_v.weight | 0xa725a500 | 0x2d0000 | +| 62 | blk.6.ffn_down.weight | 0xa752a500 | 0x6e00000 | +| 63 | blk.6.ffn_gate.weight | 0xae32a500 | 0x5a00000 | +| 64 | blk.6.ffn_norm.weight | 0xb3d2a500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0xb3d2f500 | 0x5a00000 | +| 66 | blk.7.attn_k.weight | 0xb972f500 | 0x2d0000 | +| 67 | blk.7.attn_norm.weight | 0xb99ff500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0xb9a04500 | 0xdc0000 | +| 69 | blk.7.attn_q.weight | 0xba7c4500 | 0xb40000 | +| 70 | blk.7.attn_v.weight | 0xbb304500 | 0x2d0000 | +| 71 | blk.7.ffn_down.weight | 0xbb5d4500 | 0x6e00000 | +| 72 | blk.7.ffn_gate.weight | 0xc23d4500 | 0x5a00000 | +| 73 | blk.7.ffn_norm.weight | 0xc7dd4500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0xc7dd9500 | 0x5a00000 | +| 75 | blk.8.attn_k.weight | 0xcd7d9500 | 0x2d0000 | +| 76 | blk.8.attn_norm.weight | 0xcdaa9500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0xcdaae500 | 0xdc0000 | +| 78 | blk.8.attn_q.weight | 0xce86e500 | 0xb40000 | +| 79 | blk.8.attn_v.weight | 0xcf3ae500 | 0x2d0000 | +| 80 | blk.8.ffn_down.weight | 0xcf67e500 | 0x6e00000 | +| 81 | blk.8.ffn_gate.weight | 0xd647e500 | 0x5a00000 | +| 82 | blk.8.ffn_norm.weight | 0xdbe7e500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0xdbe83500 | 0x5a00000 | +| 84 | blk.9.attn_k.weight | 0xe1883500 | 0x2d0000 | +| 85 | blk.9.attn_norm.weight | 0xe1b53500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0xe1b58500 | 0xdc0000 | +| 87 | blk.9.attn_q.weight | 0xe2918500 | 0xb40000 | +| 88 | blk.9.attn_v.weight | 0xe3458500 | 0x2d0000 | +| 89 | blk.9.ffn_down.weight | 0xe3728500 | 0x6e00000 | +| 90 | blk.9.ffn_gate.weight | 0xea528500 | 0x5a00000 | +| 91 | blk.9.ffn_norm.weight | 0xeff28500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0xeff2d500 | 0x5a00000 | +| 93 | blk.10.attn_k.weight | 0xf592d500 | 0x2d0000 | +| 94 | blk.10.attn_norm.weight | 0xf5bfd500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0xf5c02500 | 0xdc0000 | +| 96 | blk.10.attn_q.weight | 0xf69c2500 | 0xb40000 | +| 97 | blk.10.attn_v.weight | 0xf7502500 | 0x2d0000 | +| 98 | blk.10.ffn_down.weight | 0xf77d2500 | 0x6e00000 | +| 99 | blk.10.ffn_gate.weight | 0xfe5d2500 | 0x5a00000 | +| 100 | blk.10.ffn_norm.weight | 0x103fd2500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0x103fd7500 | 0x5a00000 | +| 102 | blk.11.attn_k.weight | 0x1099d7500 | 0x2d0000 | +| 103 | blk.11.attn_norm.weight | 0x109ca7500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0x109cac500 | 0xdc0000 | +| 105 | blk.11.attn_q.weight | 0x10aa6c500 | 0xb40000 | +| 106 | blk.11.attn_v.weight | 0x10b5ac500 | 0x2d0000 | +| 107 | blk.11.ffn_down.weight | 0x10b87c500 | 0x6e00000 | +| 108 | blk.11.ffn_gate.weight | 0x11267c500 | 0x5a00000 | +| 109 | blk.11.ffn_norm.weight | 0x11807c500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0x118081500 | 0x5a00000 | +| 111 | blk.12.attn_k.weight | 0x11da81500 | 0x2d0000 | +| 112 | blk.12.attn_norm.weight | 0x11dd51500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0x11dd56500 | 0xdc0000 | +| 114 | blk.12.attn_q.weight | 0x11eb16500 | 0xb40000 | +| 115 | blk.12.attn_v.weight | 0x11f656500 | 0x2d0000 | +| 116 | blk.12.ffn_down.weight | 0x11f926500 | 0x6e00000 | +| 117 | blk.12.ffn_gate.weight | 0x126726500 | 0x5a00000 | +| 118 | blk.12.ffn_norm.weight | 0x12c126500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0x12c12b500 | 0x5a00000 | +| 120 | blk.13.attn_k.weight | 0x131b2b500 | 0x2d0000 | +| 121 | blk.13.attn_norm.weight | 0x131dfb500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0x131e00500 | 0xdc0000 | +| 123 | blk.13.attn_q.weight | 0x132bc0500 | 0xb40000 | +| 124 | blk.13.attn_v.weight | 0x133700500 | 0x2d0000 | +| 125 | blk.13.ffn_down.weight | 0x1339d0500 | 0x6e00000 | +| 126 | blk.13.ffn_gate.weight | 0x13a7d0500 | 0x5a00000 | +| 127 | blk.13.ffn_norm.weight | 0x1401d0500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0x1401d5500 | 0x5a00000 | +| 129 | blk.14.attn_k.weight | 0x145bd5500 | 0x2d0000 | +| 130 | blk.14.attn_norm.weight | 0x145ea5500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0x145eaa500 | 0xdc0000 | +| 132 | blk.14.attn_q.weight | 0x146c6a500 | 0xb40000 | +| 133 | blk.14.attn_v.weight | 0x1477aa500 | 0x2d0000 | +| 134 | blk.14.ffn_down.weight | 0x147a7a500 | 0x6e00000 | +| 135 | blk.14.ffn_gate.weight | 0x14e87a500 | 0x5a00000 | +| 136 | blk.14.ffn_norm.weight | 0x15427a500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0x15427f500 | 0x5a00000 | +| 138 | blk.15.attn_k.weight | 0x159c7f500 | 0x2d0000 | +| 139 | blk.15.attn_norm.weight | 0x159f4f500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0x159f54500 | 0xdc0000 | +| 141 | blk.15.attn_q.weight | 0x15ad14500 | 0xb40000 | +| 142 | blk.15.attn_v.weight | 0x15b854500 | 0x2d0000 | +| 143 | blk.15.ffn_down.weight | 0x15bb24500 | 0x6e00000 | +| 144 | blk.15.ffn_gate.weight | 0x162924500 | 0x5a00000 | +| 145 | blk.15.ffn_norm.weight | 0x168324500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0x168329500 | 0x5a00000 | +| 147 | blk.16.attn_k.weight | 0x16dd29500 | 0x2d0000 | +| 148 | blk.16.attn_norm.weight | 0x16dff9500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0x16dffe500 | 0xdc0000 | +| 150 | blk.16.attn_q.weight | 0x16edbe500 | 0xb40000 | +| 151 | blk.16.attn_v.weight | 0x16f8fe500 | 0x2d0000 | +| 152 | blk.16.ffn_down.weight | 0x16fbce500 | 0x6e00000 | +| 153 | blk.16.ffn_gate.weight | 0x1769ce500 | 0x5a00000 | +| 154 | blk.16.ffn_norm.weight | 0x17c3ce500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0x17c3d3500 | 0x5a00000 | +| 156 | blk.17.attn_k.weight | 0x181dd3500 | 0x2d0000 | +| 157 | blk.17.attn_norm.weight | 0x1820a3500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0x1820a8500 | 0xdc0000 | +| 159 | blk.17.attn_q.weight | 0x182e68500 | 0xb40000 | +| 160 | blk.17.attn_v.weight | 0x1839a8500 | 0x2d0000 | +| 161 | blk.17.ffn_down.weight | 0x183c78500 | 0x6e00000 | +| 162 | blk.17.ffn_gate.weight | 0x18aa78500 | 0x5a00000 | +| 163 | blk.17.ffn_norm.weight | 0x190478500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0x19047d500 | 0x5a00000 | +| 165 | blk.18.attn_k.weight | 0x195e7d500 | 0x2d0000 | +| 166 | blk.18.attn_norm.weight | 0x19614d500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0x196152500 | 0xdc0000 | +| 168 | blk.18.attn_q.weight | 0x196f12500 | 0xb40000 | +| 169 | blk.18.attn_v.weight | 0x197a52500 | 0x2d0000 | +| 170 | blk.18.ffn_down.weight | 0x197d22500 | 0x6e00000 | +| 171 | blk.18.ffn_gate.weight | 0x19eb22500 | 0x5a00000 | +| 172 | blk.18.ffn_norm.weight | 0x1a4522500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x1a4527500 | 0x5a00000 | +| 174 | blk.19.attn_k.weight | 0x1a9f27500 | 0x2d0000 | +| 175 | blk.19.attn_norm.weight | 0x1aa1f7500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x1aa1fc500 | 0xdc0000 | +| 177 | blk.19.attn_q.weight | 0x1aafbc500 | 0xb40000 | +| 178 | blk.19.attn_v.weight | 0x1abafc500 | 0x2d0000 | +| 179 | blk.19.ffn_down.weight | 0x1abdcc500 | 0x6e00000 | +| 180 | blk.19.ffn_gate.weight | 0x1b2bcc500 | 0x5a00000 | +| 181 | blk.19.ffn_norm.weight | 0x1b85cc500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x1b85d1500 | 0x5a00000 | +| 183 | blk.20.attn_k.weight | 0x1bdfd1500 | 0x2d0000 | +| 184 | blk.20.attn_norm.weight | 0x1be2a1500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x1be2a6500 | 0xdc0000 | +| 186 | blk.20.attn_q.weight | 0x1bf066500 | 0xb40000 | +| 187 | blk.20.attn_v.weight | 0x1bfba6500 | 0x2d0000 | +| 188 | blk.20.ffn_down.weight | 0x1bfe76500 | 0x6e00000 | +| 189 | blk.20.ffn_gate.weight | 0x1c6c76500 | 0x5a00000 | +| 190 | blk.20.ffn_norm.weight | 0x1cc676500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x1cc67b500 | 0x5a00000 | +| 192 | blk.21.attn_k.weight | 0x1d207b500 | 0x2d0000 | +| 193 | blk.21.attn_norm.weight | 0x1d234b500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x1d2350500 | 0xdc0000 | +| 195 | blk.21.attn_q.weight | 0x1d3110500 | 0xb40000 | +| 196 | blk.21.attn_v.weight | 0x1d3c50500 | 0x2d0000 | +| 197 | blk.21.ffn_down.weight | 0x1d3f20500 | 0x6e00000 | +| 198 | blk.21.ffn_gate.weight | 0x1dad20500 | 0x5a00000 | +| 199 | blk.21.ffn_norm.weight | 0x1e0720500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x1e0725500 | 0x5a00000 | +| 201 | blk.22.attn_k.weight | 0x1e6125500 | 0x2d0000 | +| 202 | blk.22.attn_norm.weight | 0x1e63f5500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x1e63fa500 | 0xdc0000 | +| 204 | blk.22.attn_q.weight | 0x1e71ba500 | 0xb40000 | +| 205 | blk.22.attn_v.weight | 0x1e7cfa500 | 0x2d0000 | +| 206 | blk.22.ffn_down.weight | 0x1e7fca500 | 0x6e00000 | +| 207 | blk.22.ffn_gate.weight | 0x1eedca500 | 0x5a00000 | +| 208 | blk.22.ffn_norm.weight | 0x1f47ca500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x1f47cf500 | 0x5a00000 | +| 210 | blk.23.attn_k.weight | 0x1fa1cf500 | 0x2d0000 | +| 211 | blk.23.attn_norm.weight | 0x1fa49f500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x1fa4a4500 | 0xdc0000 | +| 213 | blk.23.attn_q.weight | 0x1fb264500 | 0xb40000 | +| 214 | blk.23.attn_v.weight | 0x1fbda4500 | 0x2d0000 | +| 215 | blk.23.ffn_down.weight | 0x1fc074500 | 0x6e00000 | +| 216 | blk.23.ffn_gate.weight | 0x202e74500 | 0x5a00000 | +| 217 | blk.23.ffn_norm.weight | 0x208874500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x208879500 | 0x5a00000 | +| 219 | blk.24.attn_k.weight | 0x20e279500 | 0x2d0000 | +| 220 | blk.24.attn_norm.weight | 0x20e549500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x20e54e500 | 0xdc0000 | +| 222 | blk.24.attn_q.weight | 0x20f30e500 | 0xb40000 | +| 223 | blk.24.attn_v.weight | 0x20fe4e500 | 0x2d0000 | +| 224 | blk.24.ffn_down.weight | 0x21011e500 | 0x6e00000 | +| 225 | blk.24.ffn_gate.weight | 0x216f1e500 | 0x5a00000 | +| 226 | blk.24.ffn_norm.weight | 0x21c91e500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x21c923500 | 0x5a00000 | +| 228 | blk.25.attn_k.weight | 0x222323500 | 0x370000 | +| 229 | blk.25.attn_norm.weight | 0x222693500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x222698500 | 0xdc0000 | +| 231 | blk.25.attn_q.weight | 0x223458500 | 0xdc0000 | +| 232 | blk.25.attn_v.weight | 0x224218500 | 0x370000 | +| 233 | blk.25.ffn_down.weight | 0x224588500 | 0x6e00000 | +| 234 | blk.25.ffn_gate.weight | 0x22b388500 | 0x5a00000 | +| 235 | blk.25.ffn_norm.weight | 0x230d88500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x230d8d500 | 0x5a00000 | +| 237 | blk.26.attn_k.weight | 0x23678d500 | 0x2d0000 | +| 238 | blk.26.attn_norm.weight | 0x236a5d500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x236a62500 | 0xdc0000 | +| 240 | blk.26.attn_q.weight | 0x237822500 | 0xb40000 | +| 241 | blk.26.attn_v.weight | 0x238362500 | 0x2d0000 | +| 242 | blk.26.ffn_down.weight | 0x238632500 | 0x6e00000 | +| 243 | blk.26.ffn_gate.weight | 0x23f432500 | 0x5a00000 | +| 244 | blk.26.ffn_norm.weight | 0x244e32500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x244e37500 | 0x5a00000 | +| 246 | blk.27.attn_k.weight | 0x24a837500 | 0x2d0000 | +| 247 | blk.27.attn_norm.weight | 0x24ab07500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x24ab0c500 | 0xdc0000 | +| 249 | blk.27.attn_q.weight | 0x24b8cc500 | 0xb40000 | +| 250 | blk.27.attn_v.weight | 0x24c40c500 | 0x2d0000 | +| 251 | blk.27.ffn_down.weight | 0x24c6dc500 | 0x6e00000 | +| 252 | blk.27.ffn_gate.weight | 0x2534dc500 | 0x5a00000 | +| 253 | blk.27.ffn_norm.weight | 0x258edc500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x258ee1500 | 0x5a00000 | +| 255 | blk.28.attn_k.weight | 0x25e8e1500 | 0x370000 | +| 256 | blk.28.attn_norm.weight | 0x25ec51500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x25ec56500 | 0xdc0000 | +| 258 | blk.28.attn_q.weight | 0x25fa16500 | 0xdc0000 | +| 259 | blk.28.attn_v.weight | 0x2607d6500 | 0x370000 | +| 260 | blk.28.ffn_down.weight | 0x260b46500 | 0x6e00000 | +| 261 | blk.28.ffn_gate.weight | 0x267946500 | 0x5a00000 | +| 262 | blk.28.ffn_norm.weight | 0x26d346500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x26d34b500 | 0x5a00000 | +| 264 | blk.29.attn_k.weight | 0x272d4b500 | 0x2d0000 | +| 265 | blk.29.attn_norm.weight | 0x27301b500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x273020500 | 0xdc0000 | +| 267 | blk.29.attn_q.weight | 0x273de0500 | 0xb40000 | +| 268 | blk.29.attn_v.weight | 0x274920500 | 0x2d0000 | +| 269 | blk.29.ffn_down.weight | 0x274bf0500 | 0x6e00000 | +| 270 | blk.29.ffn_gate.weight | 0x27b9f0500 | 0x5a00000 | +| 271 | blk.29.ffn_norm.weight | 0x2813f0500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x2813f5500 | 0x5a00000 | +| 273 | blk.30.attn_k.weight | 0x286df5500 | 0x370000 | +| 274 | blk.30.attn_norm.weight | 0x287165500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x28716a500 | 0xdc0000 | +| 276 | blk.30.attn_q.weight | 0x287f2a500 | 0xdc0000 | +| 277 | blk.30.attn_v.weight | 0x288cea500 | 0x370000 | +| 278 | blk.30.ffn_down.weight | 0x28905a500 | 0x6e00000 | +| 279 | blk.30.ffn_gate.weight | 0x28fe5a500 | 0x6e00000 | +| 280 | blk.30.ffn_norm.weight | 0x296c5a500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x296c5f500 | 0x6e00000 | +| 282 | blk.31.attn_k.weight | 0x29da5f500 | 0x2d0000 | +| 283 | blk.31.attn_norm.weight | 0x29dd2f500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x29dd34500 | 0xdc0000 | +| 285 | blk.31.attn_q.weight | 0x29eaf4500 | 0xb40000 | +| 286 | blk.31.attn_v.weight | 0x29f634500 | 0x2d0000 | +| 287 | blk.31.ffn_down.weight | 0x29f904500 | 0x6e00000 | +| 288 | blk.31.ffn_gate.weight | 0x2a6704500 | 0x6e00000 | +| 289 | blk.31.ffn_norm.weight | 0x2ad504500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x2ad509500 | 0x6e00000 | +| 291 | blk.32.attn_k.weight | 0x2b4309500 | 0x370000 | +| 292 | blk.32.attn_norm.weight | 0x2b4679500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x2b467e500 | 0xdc0000 | +| 294 | blk.32.attn_q.weight | 0x2b543e500 | 0xdc0000 | +| 295 | blk.32.attn_v.weight | 0x2b61fe500 | 0x370000 | +| 296 | blk.32.ffn_down.weight | 0x2b656e500 | 0x6e00000 | +| 297 | blk.32.ffn_gate.weight | 0x2bd36e500 | 0x6e00000 | +| 298 | blk.32.ffn_norm.weight | 0x2c416e500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x2c4173500 | 0x6e00000 | +| 300 | blk.33.attn_k.weight | 0x2caf73500 | 0x370000 | +| 301 | blk.33.attn_norm.weight | 0x2cb2e3500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x2cb2e8500 | 0xdc0000 | +| 303 | blk.33.attn_q.weight | 0x2cc0a8500 | 0xdc0000 | +| 304 | blk.33.attn_v.weight | 0x2cce68500 | 0x370000 | +| 305 | blk.33.ffn_down.weight | 0x2cd1d8500 | 0x6e00000 | +| 306 | blk.33.ffn_gate.weight | 0x2d3fd8500 | 0x6e00000 | +| 307 | blk.33.ffn_norm.weight | 0x2dadd8500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x2daddd500 | 0x6e00000 | +| 309 | blk.34.attn_k.weight | 0x2e1bdd500 | 0x370000 | +| 310 | blk.34.attn_norm.weight | 0x2e1f4d500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x2e1f52500 | 0xdc0000 | +| 312 | blk.34.attn_q.weight | 0x2e2d12500 | 0xdc0000 | +| 313 | blk.34.attn_v.weight | 0x2e3ad2500 | 0x370000 | +| 314 | blk.34.ffn_down.weight | 0x2e3e42500 | 0x6e00000 | +| 315 | blk.34.ffn_gate.weight | 0x2eac42500 | 0x6e00000 | +| 316 | blk.34.ffn_norm.weight | 0x2f1a42500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x2f1a47500 | 0x6e00000 | +| 318 | blk.35.attn_k.weight | 0x2f8847500 | 0x370000 | +| 319 | blk.35.attn_norm.weight | 0x2f8bb7500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x2f8bbc500 | 0xdc0000 | +| 321 | blk.35.attn_q.weight | 0x2f997c500 | 0xdc0000 | +| 322 | blk.35.attn_v.weight | 0x2fa73c500 | 0x370000 | +| 323 | blk.35.ffn_down.weight | 0x2faaac500 | 0x6e00000 | +| 324 | blk.35.ffn_gate.weight | 0x3018ac500 | 0x6e00000 | +| 325 | blk.35.ffn_norm.weight | 0x3086ac500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x3086b1500 | 0x6e00000 | +| 327 | blk.36.attn_k.weight | 0x30f4b1500 | 0x370000 | +| 328 | blk.36.attn_norm.weight | 0x30f821500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x30f826500 | 0xdc0000 | +| 330 | blk.36.attn_q.weight | 0x3105e6500 | 0xdc0000 | +| 331 | blk.36.attn_v.weight | 0x3113a6500 | 0x370000 | +| 332 | blk.36.ffn_down.weight | 0x311716500 | 0x6e00000 | +| 333 | blk.36.ffn_gate.weight | 0x318516500 | 0x6e00000 | +| 334 | blk.36.ffn_norm.weight | 0x31f316500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x31f31b500 | 0x6e00000 | +| 336 | blk.37.attn_k.weight | 0x32611b500 | 0x370000 | +| 337 | blk.37.attn_norm.weight | 0x32648b500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x326490500 | 0xdc0000 | +| 339 | blk.37.attn_q.weight | 0x327250500 | 0xdc0000 | +| 340 | blk.37.attn_v.weight | 0x328010500 | 0x370000 | +| 341 | blk.37.ffn_down.weight | 0x328380500 | 0x6e00000 | +| 342 | blk.37.ffn_gate.weight | 0x32f180500 | 0x6e00000 | +| 343 | blk.37.ffn_norm.weight | 0x335f80500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x335f85500 | 0x6e00000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q5_K | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q4_K | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q4_K | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q5_K | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q5_K | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-Q6_K.md b/scores/Dolphin-Mistral-24B-Venice-Edition-Q6_K.md new file mode 100644 index 0000000..c72ab24 --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-Q6_K.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-Q6_K.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 18 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-Q6\_K.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-q6_kgguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0x20d00000 | +| 1 | output_norm.weight | 0x21484500 | 0x5000 | +| 2 | token_embd.weight | 0x21489500 | 0x11300000 | +| 3 | blk.0.attn_k.weight | 0x32789500 | 0x370000 | +| 4 | blk.0.attn_norm.weight | 0x32af9500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x32afe500 | 0x1068000 | +| 6 | blk.0.attn_q.weight | 0x33b66500 | 0xdc0000 | +| 7 | blk.0.attn_v.weight | 0x34926500 | 0x550000 | +| 8 | blk.0.ffn_down.weight | 0x34e76500 | 0x8340000 | +| 9 | blk.0.ffn_gate.weight | 0x3d1b6500 | 0x6e00000 | +| 10 | blk.0.ffn_norm.weight | 0x43fb6500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x43fbb500 | 0x6e00000 | +| 12 | blk.1.attn_k.weight | 0x4adbb500 | 0x370000 | +| 13 | blk.1.attn_norm.weight | 0x4b12b500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x4b130500 | 0x1068000 | +| 15 | blk.1.attn_q.weight | 0x4c198500 | 0xdc0000 | +| 16 | blk.1.attn_v.weight | 0x4cf58500 | 0x550000 | +| 17 | blk.1.ffn_down.weight | 0x4d4a8500 | 0x8340000 | +| 18 | blk.1.ffn_gate.weight | 0x557e8500 | 0x6e00000 | +| 19 | blk.1.ffn_norm.weight | 0x5c5e8500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x5c5ed500 | 0x6e00000 | +| 21 | blk.2.attn_k.weight | 0x633ed500 | 0x370000 | +| 22 | blk.2.attn_norm.weight | 0x6375d500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x63762500 | 0x1068000 | +| 24 | blk.2.attn_q.weight | 0x647ca500 | 0xdc0000 | +| 25 | blk.2.attn_v.weight | 0x6558a500 | 0x550000 | +| 26 | blk.2.ffn_down.weight | 0x65ada500 | 0x8340000 | +| 27 | blk.2.ffn_gate.weight | 0x6de1a500 | 0x6e00000 | +| 28 | blk.2.ffn_norm.weight | 0x74c1a500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x74c1f500 | 0x6e00000 | +| 30 | blk.3.attn_k.weight | 0x7ba1f500 | 0x370000 | +| 31 | blk.3.attn_norm.weight | 0x7bd8f500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x7bd94500 | 0x1068000 | +| 33 | blk.3.attn_q.weight | 0x7cdfc500 | 0xdc0000 | +| 34 | blk.3.attn_v.weight | 0x7dbbc500 | 0x550000 | +| 35 | blk.3.ffn_down.weight | 0x7e10c500 | 0x8340000 | +| 36 | blk.3.ffn_gate.weight | 0x8644c500 | 0x6e00000 | +| 37 | blk.3.ffn_norm.weight | 0x8d24c500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0x8d251500 | 0x6e00000 | +| 39 | blk.4.attn_k.weight | 0x94051500 | 0x370000 | +| 40 | blk.4.attn_norm.weight | 0x943c1500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0x943c6500 | 0x1068000 | +| 42 | blk.4.attn_q.weight | 0x9542e500 | 0xdc0000 | +| 43 | blk.4.attn_v.weight | 0x961ee500 | 0x550000 | +| 44 | blk.4.ffn_down.weight | 0x9673e500 | 0x8340000 | +| 45 | blk.4.ffn_gate.weight | 0x9ea7e500 | 0x6e00000 | +| 46 | blk.4.ffn_norm.weight | 0xa587e500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0xa5883500 | 0x6e00000 | +| 48 | blk.5.attn_k.weight | 0xac683500 | 0x370000 | +| 49 | blk.5.attn_norm.weight | 0xac9f3500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0xac9f8500 | 0x1068000 | +| 51 | blk.5.attn_q.weight | 0xada60500 | 0xdc0000 | +| 52 | blk.5.attn_v.weight | 0xae820500 | 0x550000 | +| 53 | blk.5.ffn_down.weight | 0xaed70500 | 0x8340000 | +| 54 | blk.5.ffn_gate.weight | 0xb70b0500 | 0x6e00000 | +| 55 | blk.5.ffn_norm.weight | 0xbdeb0500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0xbdeb5500 | 0x6e00000 | +| 57 | blk.6.attn_k.weight | 0xc4cb5500 | 0x370000 | +| 58 | blk.6.attn_norm.weight | 0xc5025500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0xc502a500 | 0x1068000 | +| 60 | blk.6.attn_q.weight | 0xc6092500 | 0xdc0000 | +| 61 | blk.6.attn_v.weight | 0xc6e52500 | 0x550000 | +| 62 | blk.6.ffn_down.weight | 0xc73a2500 | 0x8340000 | +| 63 | blk.6.ffn_gate.weight | 0xcf6e2500 | 0x6e00000 | +| 64 | blk.6.ffn_norm.weight | 0xd64e2500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0xd64e7500 | 0x6e00000 | +| 66 | blk.7.attn_k.weight | 0xdd2e7500 | 0x370000 | +| 67 | blk.7.attn_norm.weight | 0xdd657500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0xdd65c500 | 0x1068000 | +| 69 | blk.7.attn_q.weight | 0xde6c4500 | 0xdc0000 | +| 70 | blk.7.attn_v.weight | 0xdf484500 | 0x550000 | +| 71 | blk.7.ffn_down.weight | 0xdf9d4500 | 0x8340000 | +| 72 | blk.7.ffn_gate.weight | 0xe7d14500 | 0x6e00000 | +| 73 | blk.7.ffn_norm.weight | 0xeeb14500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0xeeb19500 | 0x6e00000 | +| 75 | blk.8.attn_k.weight | 0xf5919500 | 0x370000 | +| 76 | blk.8.attn_norm.weight | 0xf5c89500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0xf5c8e500 | 0x1068000 | +| 78 | blk.8.attn_q.weight | 0xf6cf6500 | 0xdc0000 | +| 79 | blk.8.attn_v.weight | 0xf7ab6500 | 0x550000 | +| 80 | blk.8.ffn_down.weight | 0xf8006500 | 0x8340000 | +| 81 | blk.8.ffn_gate.weight | 0x100346500 | 0x6e00000 | +| 82 | blk.8.ffn_norm.weight | 0x107146500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0x10714b500 | 0x6e00000 | +| 84 | blk.9.attn_k.weight | 0x10df4b500 | 0x370000 | +| 85 | blk.9.attn_norm.weight | 0x10e2bb500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0x10e2c0500 | 0x1068000 | +| 87 | blk.9.attn_q.weight | 0x10f328500 | 0xdc0000 | +| 88 | blk.9.attn_v.weight | 0x1100e8500 | 0x550000 | +| 89 | blk.9.ffn_down.weight | 0x110638500 | 0x8340000 | +| 90 | blk.9.ffn_gate.weight | 0x118978500 | 0x6e00000 | +| 91 | blk.9.ffn_norm.weight | 0x11f778500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0x11f77d500 | 0x6e00000 | +| 93 | blk.10.attn_k.weight | 0x12657d500 | 0x370000 | +| 94 | blk.10.attn_norm.weight | 0x1268ed500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0x1268f2500 | 0x1068000 | +| 96 | blk.10.attn_q.weight | 0x12795a500 | 0xdc0000 | +| 97 | blk.10.attn_v.weight | 0x12871a500 | 0x550000 | +| 98 | blk.10.ffn_down.weight | 0x128c6a500 | 0x8340000 | +| 99 | blk.10.ffn_gate.weight | 0x130faa500 | 0x6e00000 | +| 100 | blk.10.ffn_norm.weight | 0x137daa500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0x137daf500 | 0x6e00000 | +| 102 | blk.11.attn_k.weight | 0x13ebaf500 | 0x370000 | +| 103 | blk.11.attn_norm.weight | 0x13ef1f500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0x13ef24500 | 0x1068000 | +| 105 | blk.11.attn_q.weight | 0x13ff8c500 | 0xdc0000 | +| 106 | blk.11.attn_v.weight | 0x140d4c500 | 0x550000 | +| 107 | blk.11.ffn_down.weight | 0x14129c500 | 0x8340000 | +| 108 | blk.11.ffn_gate.weight | 0x1495dc500 | 0x6e00000 | +| 109 | blk.11.ffn_norm.weight | 0x1503dc500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0x1503e1500 | 0x6e00000 | +| 111 | blk.12.attn_k.weight | 0x1571e1500 | 0x370000 | +| 112 | blk.12.attn_norm.weight | 0x157551500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0x157556500 | 0x1068000 | +| 114 | blk.12.attn_q.weight | 0x1585be500 | 0xdc0000 | +| 115 | blk.12.attn_v.weight | 0x15937e500 | 0x550000 | +| 116 | blk.12.ffn_down.weight | 0x1598ce500 | 0x8340000 | +| 117 | blk.12.ffn_gate.weight | 0x161c0e500 | 0x6e00000 | +| 118 | blk.12.ffn_norm.weight | 0x168a0e500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0x168a13500 | 0x6e00000 | +| 120 | blk.13.attn_k.weight | 0x16f813500 | 0x370000 | +| 121 | blk.13.attn_norm.weight | 0x16fb83500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0x16fb88500 | 0x1068000 | +| 123 | blk.13.attn_q.weight | 0x170bf0500 | 0xdc0000 | +| 124 | blk.13.attn_v.weight | 0x1719b0500 | 0x550000 | +| 125 | blk.13.ffn_down.weight | 0x171f00500 | 0x8340000 | +| 126 | blk.13.ffn_gate.weight | 0x17a240500 | 0x6e00000 | +| 127 | blk.13.ffn_norm.weight | 0x181040500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0x181045500 | 0x6e00000 | +| 129 | blk.14.attn_k.weight | 0x187e45500 | 0x370000 | +| 130 | blk.14.attn_norm.weight | 0x1881b5500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0x1881ba500 | 0x1068000 | +| 132 | blk.14.attn_q.weight | 0x189222500 | 0xdc0000 | +| 133 | blk.14.attn_v.weight | 0x189fe2500 | 0x550000 | +| 134 | blk.14.ffn_down.weight | 0x18a532500 | 0x8340000 | +| 135 | blk.14.ffn_gate.weight | 0x192872500 | 0x6e00000 | +| 136 | blk.14.ffn_norm.weight | 0x199672500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0x199677500 | 0x6e00000 | +| 138 | blk.15.attn_k.weight | 0x1a0477500 | 0x370000 | +| 139 | blk.15.attn_norm.weight | 0x1a07e7500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0x1a07ec500 | 0x1068000 | +| 141 | blk.15.attn_q.weight | 0x1a1854500 | 0xdc0000 | +| 142 | blk.15.attn_v.weight | 0x1a2614500 | 0x550000 | +| 143 | blk.15.ffn_down.weight | 0x1a2b64500 | 0x8340000 | +| 144 | blk.15.ffn_gate.weight | 0x1aaea4500 | 0x6e00000 | +| 145 | blk.15.ffn_norm.weight | 0x1b1ca4500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0x1b1ca9500 | 0x6e00000 | +| 147 | blk.16.attn_k.weight | 0x1b8aa9500 | 0x370000 | +| 148 | blk.16.attn_norm.weight | 0x1b8e19500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0x1b8e1e500 | 0x1068000 | +| 150 | blk.16.attn_q.weight | 0x1b9e86500 | 0xdc0000 | +| 151 | blk.16.attn_v.weight | 0x1bac46500 | 0x550000 | +| 152 | blk.16.ffn_down.weight | 0x1bb196500 | 0x8340000 | +| 153 | blk.16.ffn_gate.weight | 0x1c34d6500 | 0x6e00000 | +| 154 | blk.16.ffn_norm.weight | 0x1ca2d6500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0x1ca2db500 | 0x6e00000 | +| 156 | blk.17.attn_k.weight | 0x1d10db500 | 0x370000 | +| 157 | blk.17.attn_norm.weight | 0x1d144b500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0x1d1450500 | 0x1068000 | +| 159 | blk.17.attn_q.weight | 0x1d24b8500 | 0xdc0000 | +| 160 | blk.17.attn_v.weight | 0x1d3278500 | 0x550000 | +| 161 | blk.17.ffn_down.weight | 0x1d37c8500 | 0x8340000 | +| 162 | blk.17.ffn_gate.weight | 0x1dbb08500 | 0x6e00000 | +| 163 | blk.17.ffn_norm.weight | 0x1e2908500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0x1e290d500 | 0x6e00000 | +| 165 | blk.18.attn_k.weight | 0x1e970d500 | 0x370000 | +| 166 | blk.18.attn_norm.weight | 0x1e9a7d500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0x1e9a82500 | 0x1068000 | +| 168 | blk.18.attn_q.weight | 0x1eaaea500 | 0xdc0000 | +| 169 | blk.18.attn_v.weight | 0x1eb8aa500 | 0x550000 | +| 170 | blk.18.ffn_down.weight | 0x1ebdfa500 | 0x8340000 | +| 171 | blk.18.ffn_gate.weight | 0x1f413a500 | 0x6e00000 | +| 172 | blk.18.ffn_norm.weight | 0x1faf3a500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x1faf3f500 | 0x6e00000 | +| 174 | blk.19.attn_k.weight | 0x201d3f500 | 0x370000 | +| 175 | blk.19.attn_norm.weight | 0x2020af500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x2020b4500 | 0x1068000 | +| 177 | blk.19.attn_q.weight | 0x20311c500 | 0xdc0000 | +| 178 | blk.19.attn_v.weight | 0x203edc500 | 0x550000 | +| 179 | blk.19.ffn_down.weight | 0x20442c500 | 0x8340000 | +| 180 | blk.19.ffn_gate.weight | 0x20c76c500 | 0x6e00000 | +| 181 | blk.19.ffn_norm.weight | 0x21356c500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x213571500 | 0x6e00000 | +| 183 | blk.20.attn_k.weight | 0x21a371500 | 0x370000 | +| 184 | blk.20.attn_norm.weight | 0x21a6e1500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x21a6e6500 | 0x1068000 | +| 186 | blk.20.attn_q.weight | 0x21b74e500 | 0xdc0000 | +| 187 | blk.20.attn_v.weight | 0x21c50e500 | 0x550000 | +| 188 | blk.20.ffn_down.weight | 0x21ca5e500 | 0x8340000 | +| 189 | blk.20.ffn_gate.weight | 0x224d9e500 | 0x6e00000 | +| 190 | blk.20.ffn_norm.weight | 0x22bb9e500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x22bba3500 | 0x6e00000 | +| 192 | blk.21.attn_k.weight | 0x2329a3500 | 0x370000 | +| 193 | blk.21.attn_norm.weight | 0x232d13500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x232d18500 | 0x1068000 | +| 195 | blk.21.attn_q.weight | 0x233d80500 | 0xdc0000 | +| 196 | blk.21.attn_v.weight | 0x234b40500 | 0x550000 | +| 197 | blk.21.ffn_down.weight | 0x235090500 | 0x8340000 | +| 198 | blk.21.ffn_gate.weight | 0x23d3d0500 | 0x6e00000 | +| 199 | blk.21.ffn_norm.weight | 0x2441d0500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x2441d5500 | 0x6e00000 | +| 201 | blk.22.attn_k.weight | 0x24afd5500 | 0x370000 | +| 202 | blk.22.attn_norm.weight | 0x24b345500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x24b34a500 | 0x1068000 | +| 204 | blk.22.attn_q.weight | 0x24c3b2500 | 0xdc0000 | +| 205 | blk.22.attn_v.weight | 0x24d172500 | 0x550000 | +| 206 | blk.22.ffn_down.weight | 0x24d6c2500 | 0x8340000 | +| 207 | blk.22.ffn_gate.weight | 0x255a02500 | 0x6e00000 | +| 208 | blk.22.ffn_norm.weight | 0x25c802500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x25c807500 | 0x6e00000 | +| 210 | blk.23.attn_k.weight | 0x263607500 | 0x370000 | +| 211 | blk.23.attn_norm.weight | 0x263977500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x26397c500 | 0x1068000 | +| 213 | blk.23.attn_q.weight | 0x2649e4500 | 0xdc0000 | +| 214 | blk.23.attn_v.weight | 0x2657a4500 | 0x550000 | +| 215 | blk.23.ffn_down.weight | 0x265cf4500 | 0x8340000 | +| 216 | blk.23.ffn_gate.weight | 0x26e034500 | 0x6e00000 | +| 217 | blk.23.ffn_norm.weight | 0x274e34500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x274e39500 | 0x6e00000 | +| 219 | blk.24.attn_k.weight | 0x27bc39500 | 0x370000 | +| 220 | blk.24.attn_norm.weight | 0x27bfa9500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x27bfae500 | 0x1068000 | +| 222 | blk.24.attn_q.weight | 0x27d016500 | 0xdc0000 | +| 223 | blk.24.attn_v.weight | 0x27ddd6500 | 0x550000 | +| 224 | blk.24.ffn_down.weight | 0x27e326500 | 0x8340000 | +| 225 | blk.24.ffn_gate.weight | 0x286666500 | 0x6e00000 | +| 226 | blk.24.ffn_norm.weight | 0x28d466500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x28d46b500 | 0x6e00000 | +| 228 | blk.25.attn_k.weight | 0x29426b500 | 0x41a000 | +| 229 | blk.25.attn_norm.weight | 0x294685500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x29468a500 | 0x1068000 | +| 231 | blk.25.attn_q.weight | 0x2956f2500 | 0x1068000 | +| 232 | blk.25.attn_v.weight | 0x29675a500 | 0x550000 | +| 233 | blk.25.ffn_down.weight | 0x296caa500 | 0x8340000 | +| 234 | blk.25.ffn_gate.weight | 0x29efea500 | 0x6e00000 | +| 235 | blk.25.ffn_norm.weight | 0x2a5dea500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x2a5def500 | 0x6e00000 | +| 237 | blk.26.attn_k.weight | 0x2acbef500 | 0x370000 | +| 238 | blk.26.attn_norm.weight | 0x2acf5f500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x2acf64500 | 0x1068000 | +| 240 | blk.26.attn_q.weight | 0x2adfcc500 | 0xdc0000 | +| 241 | blk.26.attn_v.weight | 0x2aed8c500 | 0x550000 | +| 242 | blk.26.ffn_down.weight | 0x2af2dc500 | 0x8340000 | +| 243 | blk.26.ffn_gate.weight | 0x2b761c500 | 0x6e00000 | +| 244 | blk.26.ffn_norm.weight | 0x2be41c500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x2be421500 | 0x6e00000 | +| 246 | blk.27.attn_k.weight | 0x2c5221500 | 0x370000 | +| 247 | blk.27.attn_norm.weight | 0x2c5591500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x2c5596500 | 0x1068000 | +| 249 | blk.27.attn_q.weight | 0x2c65fe500 | 0xdc0000 | +| 250 | blk.27.attn_v.weight | 0x2c73be500 | 0x550000 | +| 251 | blk.27.ffn_down.weight | 0x2c790e500 | 0x8340000 | +| 252 | blk.27.ffn_gate.weight | 0x2cfc4e500 | 0x6e00000 | +| 253 | blk.27.ffn_norm.weight | 0x2d6a4e500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x2d6a53500 | 0x6e00000 | +| 255 | blk.28.attn_k.weight | 0x2dd853500 | 0x41a000 | +| 256 | blk.28.attn_norm.weight | 0x2ddc6d500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x2ddc72500 | 0x1068000 | +| 258 | blk.28.attn_q.weight | 0x2decda500 | 0x1068000 | +| 259 | blk.28.attn_v.weight | 0x2dfd42500 | 0x550000 | +| 260 | blk.28.ffn_down.weight | 0x2e0292500 | 0x8340000 | +| 261 | blk.28.ffn_gate.weight | 0x2e85d2500 | 0x6e00000 | +| 262 | blk.28.ffn_norm.weight | 0x2ef3d2500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x2ef3d7500 | 0x6e00000 | +| 264 | blk.29.attn_k.weight | 0x2f61d7500 | 0x370000 | +| 265 | blk.29.attn_norm.weight | 0x2f6547500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x2f654c500 | 0x1068000 | +| 267 | blk.29.attn_q.weight | 0x2f75b4500 | 0xdc0000 | +| 268 | blk.29.attn_v.weight | 0x2f8374500 | 0x550000 | +| 269 | blk.29.ffn_down.weight | 0x2f88c4500 | 0x8340000 | +| 270 | blk.29.ffn_gate.weight | 0x300c04500 | 0x6e00000 | +| 271 | blk.29.ffn_norm.weight | 0x307a04500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x307a09500 | 0x6e00000 | +| 273 | blk.30.attn_k.weight | 0x30e809500 | 0x41a000 | +| 274 | blk.30.attn_norm.weight | 0x30ec23500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x30ec28500 | 0x1068000 | +| 276 | blk.30.attn_q.weight | 0x30fc90500 | 0x1068000 | +| 277 | blk.30.attn_v.weight | 0x310cf8500 | 0x550000 | +| 278 | blk.30.ffn_down.weight | 0x311248500 | 0x8340000 | +| 279 | blk.30.ffn_gate.weight | 0x319588500 | 0x8340000 | +| 280 | blk.30.ffn_norm.weight | 0x3218c8500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x3218cd500 | 0x8340000 | +| 282 | blk.31.attn_k.weight | 0x329c0d500 | 0x370000 | +| 283 | blk.31.attn_norm.weight | 0x329f7d500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x329f82500 | 0x1068000 | +| 285 | blk.31.attn_q.weight | 0x32afea500 | 0xdc0000 | +| 286 | blk.31.attn_v.weight | 0x32bdaa500 | 0x550000 | +| 287 | blk.31.ffn_down.weight | 0x32c2fa500 | 0x8340000 | +| 288 | blk.31.ffn_gate.weight | 0x33463a500 | 0x8340000 | +| 289 | blk.31.ffn_norm.weight | 0x33c97a500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x33c97f500 | 0x8340000 | +| 291 | blk.32.attn_k.weight | 0x344cbf500 | 0x41a000 | +| 292 | blk.32.attn_norm.weight | 0x3450d9500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x3450de500 | 0x1068000 | +| 294 | blk.32.attn_q.weight | 0x346146500 | 0x1068000 | +| 295 | blk.32.attn_v.weight | 0x3471ae500 | 0x550000 | +| 296 | blk.32.ffn_down.weight | 0x3476fe500 | 0x8340000 | +| 297 | blk.32.ffn_gate.weight | 0x34fa3e500 | 0x8340000 | +| 298 | blk.32.ffn_norm.weight | 0x357d7e500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x357d83500 | 0x8340000 | +| 300 | blk.33.attn_k.weight | 0x3600c3500 | 0x41a000 | +| 301 | blk.33.attn_norm.weight | 0x3604dd500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x3604e2500 | 0x1068000 | +| 303 | blk.33.attn_q.weight | 0x36154a500 | 0x1068000 | +| 304 | blk.33.attn_v.weight | 0x3625b2500 | 0x550000 | +| 305 | blk.33.ffn_down.weight | 0x362b02500 | 0x8340000 | +| 306 | blk.33.ffn_gate.weight | 0x36ae42500 | 0x8340000 | +| 307 | blk.33.ffn_norm.weight | 0x373182500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x373187500 | 0x8340000 | +| 309 | blk.34.attn_k.weight | 0x37b4c7500 | 0x41a000 | +| 310 | blk.34.attn_norm.weight | 0x37b8e1500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x37b8e6500 | 0x1068000 | +| 312 | blk.34.attn_q.weight | 0x37c94e500 | 0x1068000 | +| 313 | blk.34.attn_v.weight | 0x37d9b6500 | 0x550000 | +| 314 | blk.34.ffn_down.weight | 0x37df06500 | 0x8340000 | +| 315 | blk.34.ffn_gate.weight | 0x386246500 | 0x8340000 | +| 316 | blk.34.ffn_norm.weight | 0x38e586500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x38e58b500 | 0x8340000 | +| 318 | blk.35.attn_k.weight | 0x3968cb500 | 0x41a000 | +| 319 | blk.35.attn_norm.weight | 0x396ce5500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x396cea500 | 0x1068000 | +| 321 | blk.35.attn_q.weight | 0x397d52500 | 0x1068000 | +| 322 | blk.35.attn_v.weight | 0x398dba500 | 0x550000 | +| 323 | blk.35.ffn_down.weight | 0x39930a500 | 0x8340000 | +| 324 | blk.35.ffn_gate.weight | 0x3a164a500 | 0x8340000 | +| 325 | blk.35.ffn_norm.weight | 0x3a998a500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x3a998f500 | 0x8340000 | +| 327 | blk.36.attn_k.weight | 0x3b1ccf500 | 0x41a000 | +| 328 | blk.36.attn_norm.weight | 0x3b20e9500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x3b20ee500 | 0x1068000 | +| 330 | blk.36.attn_q.weight | 0x3b3156500 | 0x1068000 | +| 331 | blk.36.attn_v.weight | 0x3b41be500 | 0x550000 | +| 332 | blk.36.ffn_down.weight | 0x3b470e500 | 0x8340000 | +| 333 | blk.36.ffn_gate.weight | 0x3bca4e500 | 0x8340000 | +| 334 | blk.36.ffn_norm.weight | 0x3c4d8e500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x3c4d93500 | 0x8340000 | +| 336 | blk.37.attn_k.weight | 0x3cd0d3500 | 0x41a000 | +| 337 | blk.37.attn_norm.weight | 0x3cd4ed500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x3cd4f2500 | 0x1068000 | +| 339 | blk.37.attn_q.weight | 0x3ce55a500 | 0x1068000 | +| 340 | blk.37.attn_v.weight | 0x3cf5c2500 | 0x550000 | +| 341 | blk.37.ffn_down.weight | 0x3cfb12500 | 0x8340000 | +| 342 | blk.37.ffn_gate.weight | 0x3d7e52500 | 0x8340000 | +| 343 | blk.37.ffn_norm.weight | 0x3e0192500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x3e0197500 | 0x8340000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q6_K | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q5_K | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q5_K | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q5_K | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q6_K | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q6_K | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + + diff --git a/scores/Dolphin-Mistral-24B-Venice-Edition-Q8_0.md b/scores/Dolphin-Mistral-24B-Venice-Edition-Q8_0.md new file mode 100644 index 0000000..fc66ad0 --- /dev/null +++ b/scores/Dolphin-Mistral-24B-Venice-Edition-Q8_0.md @@ -0,0 +1,1154 @@ +# Dolphin-Mistral-24B-Venice-Edition-pruned-Q8_0.gguf - GGUF Internal File Dump + +- Endian: LITTLE endian + +## Key Value Metadata Store + +There are 46 key-value pairs in this file + +| POS | TYPE | Count | Key | Value | +|----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| +| 1 | UINT32 | 1 | GGUF.version | 3 | +| 2 | UINT64 | 1 | GGUF.tensor_count | 345 | +| 3 | UINT64 | 1 | GGUF.kv_count | 43 | +| 4 | STRING | 1 | general.architecture | `llama` | +| 5 | STRING | 1 | general.type | `model` | +| 6 | STRING | 1 | general.name | `Dolphin Mistral 24B Venice Edition` | +| 7 | STRING | 1 | general.finetune | `Venice-Edition` | +| 8 | STRING | 1 | general.basename | `Dolphin-Mistral` | +| 9 | STRING | 1 | general.size_label | `24B` | +| 10 | STRING | 1 | general.license | `apache-2.0` | +| 11 | UINT32 | 1 | general.base_model.count | 1 | +| 12 | STRING | 1 | general.base_model.0.name | `Mistral Small 24B Instruct 2501` | +| 13 | STRING | 1 | general.base_model.0.version | `2501` | +| 14 | STRING | 1 | general.base_model.0.organization | `Mistralai` | +| 15 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-24B-Instruct-2501` | +| 16 | UINT32 | 1 | llama.context_length | 32768 | +| 17 | UINT32 | 1 | llama.embedding_length | 5120 | +| 18 | UINT32 | 1 | llama.feed_forward_length | 32768 | +| 19 | UINT32 | 1 | llama.attention.head_count | 32 | +| 20 | UINT32 | 1 | llama.attention.head_count_kv | 8 | +| 21 | FLOAT32 | 1 | llama.rope.freq_base | 100000000.0 | +| 22 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | +| 23 | UINT32 | 1 | llama.attention.key_length | 128 | +| 24 | UINT32 | 1 | llama.attention.value_length | 128 | +| 25 | UINT32 | 1 | llama.vocab_size | 131072 | +| 26 | UINT32 | 1 | llama.rope.dimension_count | 128 | +| 27 | STRING | 1 | tokenizer.ggml.model | `gpt2` | +| 28 | STRING | 1 | tokenizer.ggml.pre | `tekken` | +| 29 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | +| 30 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | +| 31 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | +| 32 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | +| 33 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | +| 34 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | +| 35 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | +| 36 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | +| 37 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | +| 38 | STRING | 1 | tokenizer.chat_template | `{%- set today = strftime_now("`...` {%- endif %}{%- endfor %}` | +| 39 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | +| 40 | UINT32 | 1 | general.quantization_version | 2 | +| 41 | UINT32 | 1 | general.file_type | 7 | +| 42 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Dolphin-Mist`...`l-24B-Venice-Edition-small.dat` | +| 43 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/combined_eur_small.txt` | +| 44 | UINT32 | 1 | quantize.imatrix.entries_count | 281 | +| 45 | UINT32 | 1 | quantize.imatrix.chunks_count | 3192 | +| 46 | UINT32 | 1 | llama.block_count | 38 | + +## Tensors Overview ~22B Elements + +Total number of elements in all tensors: 22460892160 Elements + +- [Dolphin-Mistral-24B-Venice-Edition-pruned-Q8\_0.gguf - GGUF Internal File Dump](#Dolphin-Mistral-24B-Venice-Edition-pruned-q8_0gguf---gguf-internal-file-dump) + - [Key Value Metadata Store](#key-value-metadata-store) + - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) + - [Tensor Data Offset](#tensor-data-offset) + - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) + - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) + - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) + - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) + - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) + - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) + - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) + - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) + - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) + - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) + - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) + - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) + - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) + - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) + - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) + - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) + - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) + - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) + - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) + - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) + - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) + - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) + - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) + - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) + - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) + - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) + - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) + - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) + - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) + - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) + - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) + - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) + - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) + - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) + - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) + - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) + - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) + - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) + - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) + +### Tensor Data Offset + +This table contains the offset and data segment relative to start of file + +| T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | +|-----:|:--------------------------|-----------------:|-----------------:| +| 0 | output.weight | 0x784500 | 0x2a800000 | +| 1 | output_norm.weight | 0x2af84500 | 0x5000 | +| 2 | token_embd.weight | 0x2af89500 | 0x11300000 | +| 3 | blk.0.attn_k.weight | 0x3c289500 | 0x41a000 | +| 4 | blk.0.attn_norm.weight | 0x3c6a3500 | 0x5000 | +| 5 | blk.0.attn_output.weight | 0x3c6a8500 | 0x1540000 | +| 6 | blk.0.attn_q.weight | 0x3dbe8500 | 0x1068000 | +| 7 | blk.0.attn_v.weight | 0x3ec50500 | 0xa00000 | +| 8 | blk.0.ffn_down.weight | 0x3f650500 | 0xaa00000 | +| 9 | blk.0.ffn_gate.weight | 0x4a050500 | 0x8340000 | +| 10 | blk.0.ffn_norm.weight | 0x52390500 | 0x5000 | +| 11 | blk.0.ffn_up.weight | 0x52395500 | 0x8340000 | +| 12 | blk.1.attn_k.weight | 0x5a6d5500 | 0x41a000 | +| 13 | blk.1.attn_norm.weight | 0x5aaef500 | 0x5000 | +| 14 | blk.1.attn_output.weight | 0x5aaf4500 | 0x1540000 | +| 15 | blk.1.attn_q.weight | 0x5c034500 | 0x1068000 | +| 16 | blk.1.attn_v.weight | 0x5d09c500 | 0xa00000 | +| 17 | blk.1.ffn_down.weight | 0x5da9c500 | 0xaa00000 | +| 18 | blk.1.ffn_gate.weight | 0x6849c500 | 0x8340000 | +| 19 | blk.1.ffn_norm.weight | 0x707dc500 | 0x5000 | +| 20 | blk.1.ffn_up.weight | 0x707e1500 | 0x8340000 | +| 21 | blk.2.attn_k.weight | 0x78b21500 | 0x41a000 | +| 22 | blk.2.attn_norm.weight | 0x78f3b500 | 0x5000 | +| 23 | blk.2.attn_output.weight | 0x78f40500 | 0x1540000 | +| 24 | blk.2.attn_q.weight | 0x7a480500 | 0x1068000 | +| 25 | blk.2.attn_v.weight | 0x7b4e8500 | 0xa00000 | +| 26 | blk.2.ffn_down.weight | 0x7bee8500 | 0xaa00000 | +| 27 | blk.2.ffn_gate.weight | 0x868e8500 | 0x8340000 | +| 28 | blk.2.ffn_norm.weight | 0x8ec28500 | 0x5000 | +| 29 | blk.2.ffn_up.weight | 0x8ec2d500 | 0x8340000 | +| 30 | blk.3.attn_k.weight | 0x96f6d500 | 0x41a000 | +| 31 | blk.3.attn_norm.weight | 0x97387500 | 0x5000 | +| 32 | blk.3.attn_output.weight | 0x9738c500 | 0x1540000 | +| 33 | blk.3.attn_q.weight | 0x988cc500 | 0x1068000 | +| 34 | blk.3.attn_v.weight | 0x99934500 | 0xa00000 | +| 35 | blk.3.ffn_down.weight | 0x9a334500 | 0xaa00000 | +| 36 | blk.3.ffn_gate.weight | 0xa4d34500 | 0x8340000 | +| 37 | blk.3.ffn_norm.weight | 0xad074500 | 0x5000 | +| 38 | blk.3.ffn_up.weight | 0xad079500 | 0x8340000 | +| 39 | blk.4.attn_k.weight | 0xb53b9500 | 0x41a000 | +| 40 | blk.4.attn_norm.weight | 0xb57d3500 | 0x5000 | +| 41 | blk.4.attn_output.weight | 0xb57d8500 | 0x1540000 | +| 42 | blk.4.attn_q.weight | 0xb6d18500 | 0x1068000 | +| 43 | blk.4.attn_v.weight | 0xb7d80500 | 0xa00000 | +| 44 | blk.4.ffn_down.weight | 0xb8780500 | 0xaa00000 | +| 45 | blk.4.ffn_gate.weight | 0xc3180500 | 0x8340000 | +| 46 | blk.4.ffn_norm.weight | 0xcb4c0500 | 0x5000 | +| 47 | blk.4.ffn_up.weight | 0xcb4c5500 | 0x8340000 | +| 48 | blk.5.attn_k.weight | 0xd3805500 | 0x41a000 | +| 49 | blk.5.attn_norm.weight | 0xd3c1f500 | 0x5000 | +| 50 | blk.5.attn_output.weight | 0xd3c24500 | 0x1540000 | +| 51 | blk.5.attn_q.weight | 0xd5164500 | 0x1068000 | +| 52 | blk.5.attn_v.weight | 0xd61cc500 | 0xa00000 | +| 53 | blk.5.ffn_down.weight | 0xd6bcc500 | 0xaa00000 | +| 54 | blk.5.ffn_gate.weight | 0xe15cc500 | 0x8340000 | +| 55 | blk.5.ffn_norm.weight | 0xe990c500 | 0x5000 | +| 56 | blk.5.ffn_up.weight | 0xe9911500 | 0x8340000 | +| 57 | blk.6.attn_k.weight | 0xf1c51500 | 0x41a000 | +| 58 | blk.6.attn_norm.weight | 0xf206b500 | 0x5000 | +| 59 | blk.6.attn_output.weight | 0xf2070500 | 0x1540000 | +| 60 | blk.6.attn_q.weight | 0xf35b0500 | 0x1068000 | +| 61 | blk.6.attn_v.weight | 0xf4618500 | 0xa00000 | +| 62 | blk.6.ffn_down.weight | 0xf5018500 | 0xaa00000 | +| 63 | blk.6.ffn_gate.weight | 0xffa18500 | 0x8340000 | +| 64 | blk.6.ffn_norm.weight | 0x107d58500 | 0x5000 | +| 65 | blk.6.ffn_up.weight | 0x107d5d500 | 0x8340000 | +| 66 | blk.7.attn_k.weight | 0x11009d500 | 0x41a000 | +| 67 | blk.7.attn_norm.weight | 0x1104b7500 | 0x5000 | +| 68 | blk.7.attn_output.weight | 0x1104bc500 | 0x1540000 | +| 69 | blk.7.attn_q.weight | 0x1119fc500 | 0x1068000 | +| 70 | blk.7.attn_v.weight | 0x112a64500 | 0xa00000 | +| 71 | blk.7.ffn_down.weight | 0x113464500 | 0xaa00000 | +| 72 | blk.7.ffn_gate.weight | 0x11de64500 | 0x8340000 | +| 73 | blk.7.ffn_norm.weight | 0x1261a4500 | 0x5000 | +| 74 | blk.7.ffn_up.weight | 0x1261a9500 | 0x8340000 | +| 75 | blk.8.attn_k.weight | 0x12e4e9500 | 0x41a000 | +| 76 | blk.8.attn_norm.weight | 0x12e903500 | 0x5000 | +| 77 | blk.8.attn_output.weight | 0x12e908500 | 0x1540000 | +| 78 | blk.8.attn_q.weight | 0x12fe48500 | 0x1068000 | +| 79 | blk.8.attn_v.weight | 0x130eb0500 | 0xa00000 | +| 80 | blk.8.ffn_down.weight | 0x1318b0500 | 0xaa00000 | +| 81 | blk.8.ffn_gate.weight | 0x13c2b0500 | 0x8340000 | +| 82 | blk.8.ffn_norm.weight | 0x1445f0500 | 0x5000 | +| 83 | blk.8.ffn_up.weight | 0x1445f5500 | 0x8340000 | +| 84 | blk.9.attn_k.weight | 0x14c935500 | 0x41a000 | +| 85 | blk.9.attn_norm.weight | 0x14cd4f500 | 0x5000 | +| 86 | blk.9.attn_output.weight | 0x14cd54500 | 0x1540000 | +| 87 | blk.9.attn_q.weight | 0x14e294500 | 0x1068000 | +| 88 | blk.9.attn_v.weight | 0x14f2fc500 | 0xa00000 | +| 89 | blk.9.ffn_down.weight | 0x14fcfc500 | 0xaa00000 | +| 90 | blk.9.ffn_gate.weight | 0x15a6fc500 | 0x8340000 | +| 91 | blk.9.ffn_norm.weight | 0x162a3c500 | 0x5000 | +| 92 | blk.9.ffn_up.weight | 0x162a41500 | 0x8340000 | +| 93 | blk.10.attn_k.weight | 0x16ad81500 | 0x41a000 | +| 94 | blk.10.attn_norm.weight | 0x16b19b500 | 0x5000 | +| 95 | blk.10.attn_output.weight | 0x16b1a0500 | 0x1540000 | +| 96 | blk.10.attn_q.weight | 0x16c6e0500 | 0x1068000 | +| 97 | blk.10.attn_v.weight | 0x16d748500 | 0xa00000 | +| 98 | blk.10.ffn_down.weight | 0x16e148500 | 0xaa00000 | +| 99 | blk.10.ffn_gate.weight | 0x178b48500 | 0x8340000 | +| 100 | blk.10.ffn_norm.weight | 0x180e88500 | 0x5000 | +| 101 | blk.10.ffn_up.weight | 0x180e8d500 | 0x8340000 | +| 102 | blk.11.attn_k.weight | 0x1891cd500 | 0x41a000 | +| 103 | blk.11.attn_norm.weight | 0x1895e7500 | 0x5000 | +| 104 | blk.11.attn_output.weight | 0x1895ec500 | 0x1540000 | +| 105 | blk.11.attn_q.weight | 0x18ab2c500 | 0x1068000 | +| 106 | blk.11.attn_v.weight | 0x18bb94500 | 0xa00000 | +| 107 | blk.11.ffn_down.weight | 0x18c594500 | 0xaa00000 | +| 108 | blk.11.ffn_gate.weight | 0x196f94500 | 0x8340000 | +| 109 | blk.11.ffn_norm.weight | 0x19f2d4500 | 0x5000 | +| 110 | blk.11.ffn_up.weight | 0x19f2d9500 | 0x8340000 | +| 111 | blk.12.attn_k.weight | 0x1a7619500 | 0x41a000 | +| 112 | blk.12.attn_norm.weight | 0x1a7a33500 | 0x5000 | +| 113 | blk.12.attn_output.weight | 0x1a7a38500 | 0x1540000 | +| 114 | blk.12.attn_q.weight | 0x1a8f78500 | 0x1068000 | +| 115 | blk.12.attn_v.weight | 0x1a9fe0500 | 0xa00000 | +| 116 | blk.12.ffn_down.weight | 0x1aa9e0500 | 0xaa00000 | +| 117 | blk.12.ffn_gate.weight | 0x1b53e0500 | 0x8340000 | +| 118 | blk.12.ffn_norm.weight | 0x1bd720500 | 0x5000 | +| 119 | blk.12.ffn_up.weight | 0x1bd725500 | 0x8340000 | +| 120 | blk.13.attn_k.weight | 0x1c5a65500 | 0x41a000 | +| 121 | blk.13.attn_norm.weight | 0x1c5e7f500 | 0x5000 | +| 122 | blk.13.attn_output.weight | 0x1c5e84500 | 0x1540000 | +| 123 | blk.13.attn_q.weight | 0x1c73c4500 | 0x1068000 | +| 124 | blk.13.attn_v.weight | 0x1c842c500 | 0xa00000 | +| 125 | blk.13.ffn_down.weight | 0x1c8e2c500 | 0xaa00000 | +| 126 | blk.13.ffn_gate.weight | 0x1d382c500 | 0x8340000 | +| 127 | blk.13.ffn_norm.weight | 0x1dbb6c500 | 0x5000 | +| 128 | blk.13.ffn_up.weight | 0x1dbb71500 | 0x8340000 | +| 129 | blk.14.attn_k.weight | 0x1e3eb1500 | 0x41a000 | +| 130 | blk.14.attn_norm.weight | 0x1e42cb500 | 0x5000 | +| 131 | blk.14.attn_output.weight | 0x1e42d0500 | 0x1540000 | +| 132 | blk.14.attn_q.weight | 0x1e5810500 | 0x1068000 | +| 133 | blk.14.attn_v.weight | 0x1e6878500 | 0xa00000 | +| 134 | blk.14.ffn_down.weight | 0x1e7278500 | 0xaa00000 | +| 135 | blk.14.ffn_gate.weight | 0x1f1c78500 | 0x8340000 | +| 136 | blk.14.ffn_norm.weight | 0x1f9fb8500 | 0x5000 | +| 137 | blk.14.ffn_up.weight | 0x1f9fbd500 | 0x8340000 | +| 138 | blk.15.attn_k.weight | 0x2022fd500 | 0x41a000 | +| 139 | blk.15.attn_norm.weight | 0x202717500 | 0x5000 | +| 140 | blk.15.attn_output.weight | 0x20271c500 | 0x1540000 | +| 141 | blk.15.attn_q.weight | 0x203c5c500 | 0x1068000 | +| 142 | blk.15.attn_v.weight | 0x204cc4500 | 0xa00000 | +| 143 | blk.15.ffn_down.weight | 0x2056c4500 | 0xaa00000 | +| 144 | blk.15.ffn_gate.weight | 0x2100c4500 | 0x8340000 | +| 145 | blk.15.ffn_norm.weight | 0x218404500 | 0x5000 | +| 146 | blk.15.ffn_up.weight | 0x218409500 | 0x8340000 | +| 147 | blk.16.attn_k.weight | 0x220749500 | 0x41a000 | +| 148 | blk.16.attn_norm.weight | 0x220b63500 | 0x5000 | +| 149 | blk.16.attn_output.weight | 0x220b68500 | 0x1540000 | +| 150 | blk.16.attn_q.weight | 0x2220a8500 | 0x1068000 | +| 151 | blk.16.attn_v.weight | 0x223110500 | 0xa00000 | +| 152 | blk.16.ffn_down.weight | 0x223b10500 | 0xaa00000 | +| 153 | blk.16.ffn_gate.weight | 0x22e510500 | 0x8340000 | +| 154 | blk.16.ffn_norm.weight | 0x236850500 | 0x5000 | +| 155 | blk.16.ffn_up.weight | 0x236855500 | 0x8340000 | +| 156 | blk.17.attn_k.weight | 0x23eb95500 | 0x550000 | +| 157 | blk.17.attn_norm.weight | 0x23f0e5500 | 0x5000 | +| 158 | blk.17.attn_output.weight | 0x23f0ea500 | 0x1540000 | +| 159 | blk.17.attn_q.weight | 0x24062a500 | 0x1540000 | +| 160 | blk.17.attn_v.weight | 0x241b6a500 | 0xa00000 | +| 161 | blk.17.ffn_down.weight | 0x24256a500 | 0xaa00000 | +| 162 | blk.17.ffn_gate.weight | 0x24cf6a500 | 0x8340000 | +| 163 | blk.17.ffn_norm.weight | 0x2552aa500 | 0x5000 | +| 164 | blk.17.ffn_up.weight | 0x2552af500 | 0x8340000 | +| 165 | blk.18.attn_k.weight | 0x25d5ef500 | 0x550000 | +| 166 | blk.18.attn_norm.weight | 0x25db3f500 | 0x5000 | +| 167 | blk.18.attn_output.weight | 0x25db44500 | 0x1540000 | +| 168 | blk.18.attn_q.weight | 0x25f084500 | 0x1540000 | +| 169 | blk.18.attn_v.weight | 0x2605c4500 | 0xa00000 | +| 170 | blk.18.ffn_down.weight | 0x260fc4500 | 0xaa00000 | +| 171 | blk.18.ffn_gate.weight | 0x26b9c4500 | 0x8340000 | +| 172 | blk.18.ffn_norm.weight | 0x273d04500 | 0x5000 | +| 173 | blk.18.ffn_up.weight | 0x273d09500 | 0x8340000 | +| 174 | blk.19.attn_k.weight | 0x27c049500 | 0x41a000 | +| 175 | blk.19.attn_norm.weight | 0x27c463500 | 0x5000 | +| 176 | blk.19.attn_output.weight | 0x27c468500 | 0x1540000 | +| 177 | blk.19.attn_q.weight | 0x27d9a8500 | 0x1068000 | +| 178 | blk.19.attn_v.weight | 0x27ea10500 | 0xa00000 | +| 179 | blk.19.ffn_down.weight | 0x27f410500 | 0xaa00000 | +| 180 | blk.19.ffn_gate.weight | 0x289e10500 | 0x8340000 | +| 181 | blk.19.ffn_norm.weight | 0x292150500 | 0x5000 | +| 182 | blk.19.ffn_up.weight | 0x292155500 | 0x8340000 | +| 183 | blk.20.attn_k.weight | 0x29a495500 | 0x550000 | +| 184 | blk.20.attn_norm.weight | 0x29a9e5500 | 0x5000 | +| 185 | blk.20.attn_output.weight | 0x29a9ea500 | 0x1540000 | +| 186 | blk.20.attn_q.weight | 0x29bf2a500 | 0x1540000 | +| 187 | blk.20.attn_v.weight | 0x29d46a500 | 0xa00000 | +| 188 | blk.20.ffn_down.weight | 0x29de6a500 | 0xaa00000 | +| 189 | blk.20.ffn_gate.weight | 0x2a886a500 | 0xaa00000 | +| 190 | blk.20.ffn_norm.weight | 0x2b326a500 | 0x5000 | +| 191 | blk.20.ffn_up.weight | 0x2b326f500 | 0xaa00000 | +| 192 | blk.21.attn_k.weight | 0x2bdc6f500 | 0x41a000 | +| 193 | blk.21.attn_norm.weight | 0x2be089500 | 0x5000 | +| 194 | blk.21.attn_output.weight | 0x2be08e500 | 0x1540000 | +| 195 | blk.21.attn_q.weight | 0x2bf5ce500 | 0x1068000 | +| 196 | blk.21.attn_v.weight | 0x2c0636500 | 0xa00000 | +| 197 | blk.21.ffn_down.weight | 0x2c1036500 | 0xaa00000 | +| 198 | blk.21.ffn_gate.weight | 0x2cba36500 | 0xaa00000 | +| 199 | blk.21.ffn_norm.weight | 0x2d6436500 | 0x5000 | +| 200 | blk.21.ffn_up.weight | 0x2d643b500 | 0xaa00000 | +| 201 | blk.22.attn_k.weight | 0x2e0e3b500 | 0x550000 | +| 202 | blk.22.attn_norm.weight | 0x2e138b500 | 0x5000 | +| 203 | blk.22.attn_output.weight | 0x2e1390500 | 0x1540000 | +| 204 | blk.22.attn_q.weight | 0x2e28d0500 | 0x1540000 | +| 205 | blk.22.attn_v.weight | 0x2e3e10500 | 0xa00000 | +| 206 | blk.22.ffn_down.weight | 0x2e4810500 | 0xaa00000 | +| 207 | blk.22.ffn_gate.weight | 0x2ef210500 | 0xaa00000 | +| 208 | blk.22.ffn_norm.weight | 0x2f9c10500 | 0x5000 | +| 209 | blk.22.ffn_up.weight | 0x2f9c15500 | 0xaa00000 | +| 210 | blk.23.attn_k.weight | 0x304615500 | 0x550000 | +| 211 | blk.23.attn_norm.weight | 0x304b65500 | 0x5000 | +| 212 | blk.23.attn_output.weight | 0x304b6a500 | 0x1540000 | +| 213 | blk.23.attn_q.weight | 0x3060aa500 | 0x1540000 | +| 214 | blk.23.attn_v.weight | 0x3075ea500 | 0xa00000 | +| 215 | blk.23.ffn_down.weight | 0x307fea500 | 0xaa00000 | +| 216 | blk.23.ffn_gate.weight | 0x3129ea500 | 0xaa00000 | +| 217 | blk.23.ffn_norm.weight | 0x31d3ea500 | 0x5000 | +| 218 | blk.23.ffn_up.weight | 0x31d3ef500 | 0xaa00000 | +| 219 | blk.24.attn_k.weight | 0x327def500 | 0x550000 | +| 220 | blk.24.attn_norm.weight | 0x32833f500 | 0x5000 | +| 221 | blk.24.attn_output.weight | 0x328344500 | 0x1540000 | +| 222 | blk.24.attn_q.weight | 0x329884500 | 0x1540000 | +| 223 | blk.24.attn_v.weight | 0x32adc4500 | 0xa00000 | +| 224 | blk.24.ffn_down.weight | 0x32b7c4500 | 0xaa00000 | +| 225 | blk.24.ffn_gate.weight | 0x3361c4500 | 0xaa00000 | +| 226 | blk.24.ffn_norm.weight | 0x340bc4500 | 0x5000 | +| 227 | blk.24.ffn_up.weight | 0x340bc9500 | 0xaa00000 | +| 228 | blk.25.attn_k.weight | 0x34b5c9500 | 0x550000 | +| 229 | blk.25.attn_norm.weight | 0x34bb19500 | 0x5000 | +| 230 | blk.25.attn_output.weight | 0x34bb1e500 | 0x1540000 | +| 231 | blk.25.attn_q.weight | 0x34d05e500 | 0x1540000 | +| 232 | blk.25.attn_v.weight | 0x34e59e500 | 0xa00000 | +| 233 | blk.25.ffn_down.weight | 0x34ef9e500 | 0xaa00000 | +| 234 | blk.25.ffn_gate.weight | 0x35999e500 | 0xaa00000 | +| 235 | blk.25.ffn_norm.weight | 0x36439e500 | 0x5000 | +| 236 | blk.25.ffn_up.weight | 0x3643a3500 | 0xaa00000 | +| 237 | blk.26.attn_k.weight | 0x36eda3500 | 0x550000 | +| 238 | blk.26.attn_norm.weight | 0x36f2f3500 | 0x5000 | +| 239 | blk.26.attn_output.weight | 0x36f2f8500 | 0x1540000 | +| 240 | blk.26.attn_q.weight | 0x370838500 | 0x1540000 | +| 241 | blk.26.attn_v.weight | 0x371d78500 | 0xa00000 | +| 242 | blk.26.ffn_down.weight | 0x372778500 | 0xaa00000 | +| 243 | blk.26.ffn_gate.weight | 0x37d178500 | 0xaa00000 | +| 244 | blk.26.ffn_norm.weight | 0x387b78500 | 0x5000 | +| 245 | blk.26.ffn_up.weight | 0x387b7d500 | 0xaa00000 | +| 246 | blk.27.attn_k.weight | 0x39257d500 | 0x41a000 | +| 247 | blk.27.attn_norm.weight | 0x392997500 | 0x5000 | +| 248 | blk.27.attn_output.weight | 0x39299c500 | 0x1540000 | +| 249 | blk.27.attn_q.weight | 0x393edc500 | 0x1068000 | +| 250 | blk.27.attn_v.weight | 0x394f44500 | 0xa00000 | +| 251 | blk.27.ffn_down.weight | 0x395944500 | 0xaa00000 | +| 252 | blk.27.ffn_gate.weight | 0x3a0344500 | 0xaa00000 | +| 253 | blk.27.ffn_norm.weight | 0x3aad44500 | 0x5000 | +| 254 | blk.27.ffn_up.weight | 0x3aad49500 | 0xaa00000 | +| 255 | blk.28.attn_k.weight | 0x3b5749500 | 0x550000 | +| 256 | blk.28.attn_norm.weight | 0x3b5c99500 | 0x5000 | +| 257 | blk.28.attn_output.weight | 0x3b5c9e500 | 0x1540000 | +| 258 | blk.28.attn_q.weight | 0x3b71de500 | 0x1540000 | +| 259 | blk.28.attn_v.weight | 0x3b871e500 | 0xa00000 | +| 260 | blk.28.ffn_down.weight | 0x3b911e500 | 0xaa00000 | +| 261 | blk.28.ffn_gate.weight | 0x3c3b1e500 | 0xaa00000 | +| 262 | blk.28.ffn_norm.weight | 0x3ce51e500 | 0x5000 | +| 263 | blk.28.ffn_up.weight | 0x3ce523500 | 0xaa00000 | +| 264 | blk.29.attn_k.weight | 0x3d8f23500 | 0x550000 | +| 265 | blk.29.attn_norm.weight | 0x3d9473500 | 0x5000 | +| 266 | blk.29.attn_output.weight | 0x3d9478500 | 0x1540000 | +| 267 | blk.29.attn_q.weight | 0x3da9b8500 | 0x1540000 | +| 268 | blk.29.attn_v.weight | 0x3dbef8500 | 0xa00000 | +| 269 | blk.29.ffn_down.weight | 0x3dc8f8500 | 0xaa00000 | +| 270 | blk.29.ffn_gate.weight | 0x3e72f8500 | 0xaa00000 | +| 271 | blk.29.ffn_norm.weight | 0x3f1cf8500 | 0x5000 | +| 272 | blk.29.ffn_up.weight | 0x3f1cfd500 | 0xaa00000 | +| 273 | blk.30.attn_k.weight | 0x3fc6fd500 | 0x550000 | +| 274 | blk.30.attn_norm.weight | 0x3fcc4d500 | 0x5000 | +| 275 | blk.30.attn_output.weight | 0x3fcc52500 | 0x1540000 | +| 276 | blk.30.attn_q.weight | 0x3fe192500 | 0x1540000 | +| 277 | blk.30.attn_v.weight | 0x3ff6d2500 | 0xa00000 | +| 278 | blk.30.ffn_down.weight | 0x4000d2500 | 0xaa00000 | +| 279 | blk.30.ffn_gate.weight | 0x40aad2500 | 0xaa00000 | +| 280 | blk.30.ffn_norm.weight | 0x4154d2500 | 0x5000 | +| 281 | blk.30.ffn_up.weight | 0x4154d7500 | 0xaa00000 | +| 282 | blk.31.attn_k.weight | 0x41fed7500 | 0x550000 | +| 283 | blk.31.attn_norm.weight | 0x420427500 | 0x5000 | +| 284 | blk.31.attn_output.weight | 0x42042c500 | 0x1540000 | +| 285 | blk.31.attn_q.weight | 0x42196c500 | 0x1540000 | +| 286 | blk.31.attn_v.weight | 0x422eac500 | 0xa00000 | +| 287 | blk.31.ffn_down.weight | 0x4238ac500 | 0xaa00000 | +| 288 | blk.31.ffn_gate.weight | 0x42e2ac500 | 0xaa00000 | +| 289 | blk.31.ffn_norm.weight | 0x438cac500 | 0x5000 | +| 290 | blk.31.ffn_up.weight | 0x438cb1500 | 0xaa00000 | +| 291 | blk.32.attn_k.weight | 0x4436b1500 | 0x550000 | +| 292 | blk.32.attn_norm.weight | 0x443c01500 | 0x5000 | +| 293 | blk.32.attn_output.weight | 0x443c06500 | 0x1540000 | +| 294 | blk.32.attn_q.weight | 0x445146500 | 0x1540000 | +| 295 | blk.32.attn_v.weight | 0x446686500 | 0xa00000 | +| 296 | blk.32.ffn_down.weight | 0x447086500 | 0xaa00000 | +| 297 | blk.32.ffn_gate.weight | 0x451a86500 | 0xaa00000 | +| 298 | blk.32.ffn_norm.weight | 0x45c486500 | 0x5000 | +| 299 | blk.32.ffn_up.weight | 0x45c48b500 | 0xaa00000 | +| 300 | blk.33.attn_k.weight | 0x466e8b500 | 0x550000 | +| 301 | blk.33.attn_norm.weight | 0x4673db500 | 0x5000 | +| 302 | blk.33.attn_output.weight | 0x4673e0500 | 0x1540000 | +| 303 | blk.33.attn_q.weight | 0x468920500 | 0x1540000 | +| 304 | blk.33.attn_v.weight | 0x469e60500 | 0xa00000 | +| 305 | blk.33.ffn_down.weight | 0x46a860500 | 0xaa00000 | +| 306 | blk.33.ffn_gate.weight | 0x475260500 | 0xaa00000 | +| 307 | blk.33.ffn_norm.weight | 0x47fc60500 | 0x5000 | +| 308 | blk.33.ffn_up.weight | 0x47fc65500 | 0xaa00000 | +| 309 | blk.34.attn_k.weight | 0x48a665500 | 0x550000 | +| 310 | blk.34.attn_norm.weight | 0x48abb5500 | 0x5000 | +| 311 | blk.34.attn_output.weight | 0x48abba500 | 0x1540000 | +| 312 | blk.34.attn_q.weight | 0x48c0fa500 | 0x1540000 | +| 313 | blk.34.attn_v.weight | 0x48d63a500 | 0xa00000 | +| 314 | blk.34.ffn_down.weight | 0x48e03a500 | 0xaa00000 | +| 315 | blk.34.ffn_gate.weight | 0x498a3a500 | 0xaa00000 | +| 316 | blk.34.ffn_norm.weight | 0x4a343a500 | 0x5000 | +| 317 | blk.34.ffn_up.weight | 0x4a343f500 | 0xaa00000 | +| 318 | blk.35.attn_k.weight | 0x4ade3f500 | 0x550000 | +| 319 | blk.35.attn_norm.weight | 0x4ae38f500 | 0x5000 | +| 320 | blk.35.attn_output.weight | 0x4ae394500 | 0x1540000 | +| 321 | blk.35.attn_q.weight | 0x4af8d4500 | 0x1540000 | +| 322 | blk.35.attn_v.weight | 0x4b0e14500 | 0xa00000 | +| 323 | blk.35.ffn_down.weight | 0x4b1814500 | 0xaa00000 | +| 324 | blk.35.ffn_gate.weight | 0x4bc214500 | 0xaa00000 | +| 325 | blk.35.ffn_norm.weight | 0x4c6c14500 | 0x5000 | +| 326 | blk.35.ffn_up.weight | 0x4c6c19500 | 0xaa00000 | +| 327 | blk.36.attn_k.weight | 0x4d1619500 | 0x550000 | +| 328 | blk.36.attn_norm.weight | 0x4d1b69500 | 0x5000 | +| 329 | blk.36.attn_output.weight | 0x4d1b6e500 | 0x1540000 | +| 330 | blk.36.attn_q.weight | 0x4d30ae500 | 0x1540000 | +| 331 | blk.36.attn_v.weight | 0x4d45ee500 | 0xa00000 | +| 332 | blk.36.ffn_down.weight | 0x4d4fee500 | 0xaa00000 | +| 333 | blk.36.ffn_gate.weight | 0x4df9ee500 | 0xaa00000 | +| 334 | blk.36.ffn_norm.weight | 0x4ea3ee500 | 0x5000 | +| 335 | blk.36.ffn_up.weight | 0x4ea3f3500 | 0xaa00000 | +| 336 | blk.37.attn_k.weight | 0x4f4df3500 | 0x550000 | +| 337 | blk.37.attn_norm.weight | 0x4f5343500 | 0x5000 | +| 338 | blk.37.attn_output.weight | 0x4f5348500 | 0x1540000 | +| 339 | blk.37.attn_q.weight | 0x4f6888500 | 0x1540000 | +| 340 | blk.37.attn_v.weight | 0x4f7dc8500 | 0xa00000 | +| 341 | blk.37.ffn_down.weight | 0x4f87c8500 | 0xaa00000 | +| 342 | blk.37.ffn_gate.weight | 0x5031c8500 | 0xaa00000 | +| 343 | blk.37.ffn_norm.weight | 0x50dbc8500 | 0x5000 | +| 344 | blk.37.ffn_up.weight | 0x50dbcd500 | 0xaa00000 | + +### Base Tensor Group : ~1B Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----| +| 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q8_0 | +| 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | + +- Total elements in base: ( ~1B) 1342182400 +- Percentage of total elements: 5.98% + + +### Block 0 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.0: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 1 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.1: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 2 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.2: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 3 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.3: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 4 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.4: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 5 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.5: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 6 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.6: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 7 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.7: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 8 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.8: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 9 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----| +| 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.9: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 10 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.10: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 11 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.11: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 12 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.12: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 13 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.13: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 14 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.14: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 15 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.15: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 16 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.16: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 17 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.17: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 18 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.18: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 19 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | +| 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q6_K | + +- Total elements in blk.19: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 20 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.20: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 21 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.21: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 22 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.22: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 23 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.23: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 24 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.24: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 25 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.25: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 26 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.26: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 27 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q6_K | +| 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q6_K | +| 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.27: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 28 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.28: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 29 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.29: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 30 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.30: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 31 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.31: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 32 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.32: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 33 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.33: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 34 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.34: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 35 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.35: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 36 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.36: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +### Block 37 Tensor Group : ~556M Elements + +| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | +|-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----| +| 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q8_0 | +| 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q8_0 | +| 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q8_0 | +| 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | F16 | +| 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q8_0 | +| 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | +| 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | +| 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q8_0 | + +- Total elements in blk.37: (~556M) 555755520 +- Percentage of total elements: 2.47% + + +