1030 lines
91 KiB
Markdown
1030 lines
91 KiB
Markdown
# Hammer2.1-7b-IQ3_M.gguf - GGUF Internal File Dump
|
|
|
|
- Endian: LITTLE endian
|
|
|
|
## Key Value Metadata Store
|
|
|
|
There are 36 key-value pairs in this file
|
|
|
|
| POS | TYPE | Count | Key | Value |
|
|
|----:|:---------|-------:|:-------------------------------------------|:--------------------------------------------------------------------|
|
|
| 1 | UINT32 | 1 | GGUF.version | 3 |
|
|
| 2 | UINT64 | 1 | GGUF.tensor_count | 339 |
|
|
| 3 | UINT64 | 1 | GGUF.kv_count | 33 |
|
|
| 4 | STRING | 1 | general.architecture | `qwen2` |
|
|
| 5 | STRING | 1 | general.type | `model` |
|
|
| 6 | STRING | 1 | general.name | `Hammer2.1 7b GGUF` |
|
|
| 7 | STRING | 1 | general.finetune | `GGUF` |
|
|
| 8 | STRING | 1 | general.basename | `Hammer2.1` |
|
|
| 9 | STRING | 1 | general.size_label | `7B` |
|
|
| 10 | UINT32 | 1 | qwen2.block_count | 28 |
|
|
| 11 | UINT32 | 1 | qwen2.context_length | 32768 |
|
|
| 12 | UINT32 | 1 | qwen2.embedding_length | 3584 |
|
|
| 13 | UINT32 | 1 | qwen2.feed_forward_length | 18944 |
|
|
| 14 | UINT32 | 1 | qwen2.attention.head_count | 28 |
|
|
| 15 | UINT32 | 1 | qwen2.attention.head_count_kv | 4 |
|
|
| 16 | FLOAT32 | 1 | qwen2.rope.freq_base | 1000000.0 |
|
|
| 17 | FLOAT32 | 1 | qwen2.attention.layer_norm_rms_epsilon | 1e-06 |
|
|
| 18 | STRING | 1 | qwen2.rope.scaling.type | `yarn` |
|
|
| 19 | FLOAT32 | 1 | qwen2.rope.scaling.factor | 4.0 |
|
|
| 20 | UINT32 | 1 | qwen2.rope.scaling.original_context_length | 32768 |
|
|
| 21 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
|
|
| 22 | STRING | 1 | tokenizer.ggml.pre | `qwen2` |
|
|
| 23 | [STRING] | 151665 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
|
|
| 24 | [INT32] | 151665 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
|
|
| 25 | [STRING] | 151387 | tokenizer.ggml.merges | [ `Ġ Ġ`, `ĠĠ ĠĠ`, `i n`, `Ġ t`, `ĠĠĠĠ ĠĠĠĠ`, ... ] |
|
|
| 26 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 151645 |
|
|
| 27 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 151643 |
|
|
| 28 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 151643 |
|
|
| 29 | BOOL | 1 | tokenizer.ggml.add_bos_token | False |
|
|
| 30 | STRING | 1 | tokenizer.chat_template | `{%- set system_message = 'You `...`{- '<|im_start|>assistant' }}` |
|
|
| 31 | UINT32 | 1 | general.quantization_version | 2 |
|
|
| 32 | UINT32 | 1 | general.file_type | 27 |
|
|
| 33 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Hammer2.1-7b-small.dat` |
|
|
| 34 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
|
|
| 35 | INT32 | 1 | quantize.imatrix.entries_count | 197 |
|
|
| 36 | INT32 | 1 | quantize.imatrix.chunks_count | 968 |
|
|
|
|
## Tensors Overview ~8B Elements
|
|
|
|
Total number of elements in all tensors: 7612756480 Elements
|
|
|
|
- [Hammer2.1-7b-IQ3\_M.gguf - GGUF Internal File Dump](#hammer21-7b-iq3_mgguf---gguf-internal-file-dump)
|
|
- [Key Value Metadata Store](#key-value-metadata-store)
|
|
- [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
|
|
- [Tensor Data Offset](#tensor-data-offset)
|
|
- [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
|
|
- [Block 0 Tensor Group : ~233M Elements](#block-0-tensor-group--233m-elements)
|
|
- [Block 1 Tensor Group : ~233M Elements](#block-1-tensor-group--233m-elements)
|
|
- [Block 2 Tensor Group : ~233M Elements](#block-2-tensor-group--233m-elements)
|
|
- [Block 3 Tensor Group : ~233M Elements](#block-3-tensor-group--233m-elements)
|
|
- [Block 4 Tensor Group : ~233M Elements](#block-4-tensor-group--233m-elements)
|
|
- [Block 5 Tensor Group : ~233M Elements](#block-5-tensor-group--233m-elements)
|
|
- [Block 6 Tensor Group : ~233M Elements](#block-6-tensor-group--233m-elements)
|
|
- [Block 7 Tensor Group : ~233M Elements](#block-7-tensor-group--233m-elements)
|
|
- [Block 8 Tensor Group : ~233M Elements](#block-8-tensor-group--233m-elements)
|
|
- [Block 9 Tensor Group : ~233M Elements](#block-9-tensor-group--233m-elements)
|
|
- [Block 10 Tensor Group : ~233M Elements](#block-10-tensor-group--233m-elements)
|
|
- [Block 11 Tensor Group : ~233M Elements](#block-11-tensor-group--233m-elements)
|
|
- [Block 12 Tensor Group : ~233M Elements](#block-12-tensor-group--233m-elements)
|
|
- [Block 13 Tensor Group : ~233M Elements](#block-13-tensor-group--233m-elements)
|
|
- [Block 14 Tensor Group : ~233M Elements](#block-14-tensor-group--233m-elements)
|
|
- [Block 15 Tensor Group : ~233M Elements](#block-15-tensor-group--233m-elements)
|
|
- [Block 16 Tensor Group : ~233M Elements](#block-16-tensor-group--233m-elements)
|
|
- [Block 17 Tensor Group : ~233M Elements](#block-17-tensor-group--233m-elements)
|
|
- [Block 18 Tensor Group : ~233M Elements](#block-18-tensor-group--233m-elements)
|
|
- [Block 19 Tensor Group : ~233M Elements](#block-19-tensor-group--233m-elements)
|
|
- [Block 20 Tensor Group : ~233M Elements](#block-20-tensor-group--233m-elements)
|
|
- [Block 21 Tensor Group : ~233M Elements](#block-21-tensor-group--233m-elements)
|
|
- [Block 22 Tensor Group : ~233M Elements](#block-22-tensor-group--233m-elements)
|
|
- [Block 23 Tensor Group : ~233M Elements](#block-23-tensor-group--233m-elements)
|
|
- [Block 24 Tensor Group : ~233M Elements](#block-24-tensor-group--233m-elements)
|
|
- [Block 25 Tensor Group : ~233M Elements](#block-25-tensor-group--233m-elements)
|
|
- [Block 26 Tensor Group : ~233M Elements](#block-26-tensor-group--233m-elements)
|
|
- [Block 27 Tensor Group : ~233M Elements](#block-27-tensor-group--233m-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 | 0x5ab4c0 | 0xdebe7c4 |
|
|
| 1 | output_norm.weight | 0xe469ca0 | 0x3800 |
|
|
| 2 | token_embd.weight | 0xe46d4a0 | 0xdebe7c4 |
|
|
| 3 | blk.0.attn_k.bias | 0x1c32bc80 | 0x800 |
|
|
| 4 | blk.0.attn_k.weight | 0x1c32c480 | 0xab800 |
|
|
| 5 | blk.0.attn_norm.weight | 0x1c3d7c80 | 0x3800 |
|
|
| 6 | blk.0.attn_output.weight | 0x1c3db480 | 0x6e4000 |
|
|
| 7 | blk.0.attn_q.bias | 0x1cabf480 | 0x3800 |
|
|
| 8 | blk.0.attn_q.weight | 0x1cac2c80 | 0x4b0800 |
|
|
| 9 | blk.0.attn_v.bias | 0x1cf73480 | 0x800 |
|
|
| 10 | blk.0.attn_v.weight | 0x1cf73c80 | 0xc0800 |
|
|
| 11 | blk.0.ffn_down.weight | 0x1d034480 | 0x246c000 |
|
|
| 12 | blk.0.ffn_gate.weight | 0x1f4a0480 | 0x18c9800 |
|
|
| 13 | blk.0.ffn_norm.weight | 0x20d69c80 | 0x3800 |
|
|
| 14 | blk.0.ffn_up.weight | 0x20d6d480 | 0x18c9800 |
|
|
| 15 | blk.1.attn_k.bias | 0x22636c80 | 0x800 |
|
|
| 16 | blk.1.attn_k.weight | 0x22637480 | 0xab800 |
|
|
| 17 | blk.1.attn_norm.weight | 0x226e2c80 | 0x3800 |
|
|
| 18 | blk.1.attn_output.weight | 0x226e6480 | 0x6e4000 |
|
|
| 19 | blk.1.attn_q.bias | 0x22dca480 | 0x3800 |
|
|
| 20 | blk.1.attn_q.weight | 0x22dcdc80 | 0x4b0800 |
|
|
| 21 | blk.1.attn_v.bias | 0x2327e480 | 0x800 |
|
|
| 22 | blk.1.attn_v.weight | 0x2327ec80 | 0xc0800 |
|
|
| 23 | blk.1.ffn_down.weight | 0x2333f480 | 0x246c000 |
|
|
| 24 | blk.1.ffn_gate.weight | 0x257ab480 | 0x1bd2800 |
|
|
| 25 | blk.1.ffn_norm.weight | 0x2737dc80 | 0x3800 |
|
|
| 26 | blk.1.ffn_up.weight | 0x27381480 | 0x1bd2800 |
|
|
| 27 | blk.2.attn_k.bias | 0x28f53c80 | 0x800 |
|
|
| 28 | blk.2.attn_k.weight | 0x28f54480 | 0xab800 |
|
|
| 29 | blk.2.attn_norm.weight | 0x28fffc80 | 0x3800 |
|
|
| 30 | blk.2.attn_output.weight | 0x29003480 | 0x6e4000 |
|
|
| 31 | blk.2.attn_q.bias | 0x296e7480 | 0x3800 |
|
|
| 32 | blk.2.attn_q.weight | 0x296eac80 | 0x4b0800 |
|
|
| 33 | blk.2.attn_v.bias | 0x29b9b480 | 0x800 |
|
|
| 34 | blk.2.attn_v.weight | 0x29b9bc80 | 0xc0800 |
|
|
| 35 | blk.2.ffn_down.weight | 0x29c5c480 | 0x246c000 |
|
|
| 36 | blk.2.ffn_gate.weight | 0x2c0c8480 | 0x1bd2800 |
|
|
| 37 | blk.2.ffn_norm.weight | 0x2dc9ac80 | 0x3800 |
|
|
| 38 | blk.2.ffn_up.weight | 0x2dc9e480 | 0x1bd2800 |
|
|
| 39 | blk.3.attn_k.bias | 0x2f870c80 | 0x800 |
|
|
| 40 | blk.3.attn_k.weight | 0x2f871480 | 0xab800 |
|
|
| 41 | blk.3.attn_norm.weight | 0x2f91cc80 | 0x3800 |
|
|
| 42 | blk.3.attn_output.weight | 0x2f920480 | 0x6e4000 |
|
|
| 43 | blk.3.attn_q.bias | 0x30004480 | 0x3800 |
|
|
| 44 | blk.3.attn_q.weight | 0x30007c80 | 0x4b0800 |
|
|
| 45 | blk.3.attn_v.bias | 0x304b8480 | 0x800 |
|
|
| 46 | blk.3.attn_v.weight | 0x304b8c80 | 0xc0800 |
|
|
| 47 | blk.3.ffn_down.weight | 0x30579480 | 0x246c000 |
|
|
| 48 | blk.3.ffn_gate.weight | 0x329e5480 | 0x1bd2800 |
|
|
| 49 | blk.3.ffn_norm.weight | 0x345b7c80 | 0x3800 |
|
|
| 50 | blk.3.ffn_up.weight | 0x345bb480 | 0x1bd2800 |
|
|
| 51 | blk.4.attn_k.bias | 0x3618dc80 | 0x800 |
|
|
| 52 | blk.4.attn_k.weight | 0x3618e480 | 0xab800 |
|
|
| 53 | blk.4.attn_norm.weight | 0x36239c80 | 0x3800 |
|
|
| 54 | blk.4.attn_output.weight | 0x3623d480 | 0x6e4000 |
|
|
| 55 | blk.4.attn_q.bias | 0x36921480 | 0x3800 |
|
|
| 56 | blk.4.attn_q.weight | 0x36924c80 | 0x4b0800 |
|
|
| 57 | blk.4.attn_v.bias | 0x36dd5480 | 0x800 |
|
|
| 58 | blk.4.attn_v.weight | 0x36dd5c80 | 0xc0800 |
|
|
| 59 | blk.4.ffn_down.weight | 0x36e96480 | 0x246c000 |
|
|
| 60 | blk.4.ffn_gate.weight | 0x39302480 | 0x1bd2800 |
|
|
| 61 | blk.4.ffn_norm.weight | 0x3aed4c80 | 0x3800 |
|
|
| 62 | blk.4.ffn_up.weight | 0x3aed8480 | 0x1bd2800 |
|
|
| 63 | blk.5.attn_k.bias | 0x3caaac80 | 0x800 |
|
|
| 64 | blk.5.attn_k.weight | 0x3caab480 | 0xab800 |
|
|
| 65 | blk.5.attn_norm.weight | 0x3cb56c80 | 0x3800 |
|
|
| 66 | blk.5.attn_output.weight | 0x3cb5a480 | 0x6e4000 |
|
|
| 67 | blk.5.attn_q.bias | 0x3d23e480 | 0x3800 |
|
|
| 68 | blk.5.attn_q.weight | 0x3d241c80 | 0x4b0800 |
|
|
| 69 | blk.5.attn_v.bias | 0x3d6f2480 | 0x800 |
|
|
| 70 | blk.5.attn_v.weight | 0x3d6f2c80 | 0xc0800 |
|
|
| 71 | blk.5.ffn_down.weight | 0x3d7b3480 | 0x246c000 |
|
|
| 72 | blk.5.ffn_gate.weight | 0x3fc1f480 | 0x1bd2800 |
|
|
| 73 | blk.5.ffn_norm.weight | 0x417f1c80 | 0x3800 |
|
|
| 74 | blk.5.ffn_up.weight | 0x417f5480 | 0x1bd2800 |
|
|
| 75 | blk.6.attn_k.bias | 0x433c7c80 | 0x800 |
|
|
| 76 | blk.6.attn_k.weight | 0x433c8480 | 0xab800 |
|
|
| 77 | blk.6.attn_norm.weight | 0x43473c80 | 0x3800 |
|
|
| 78 | blk.6.attn_output.weight | 0x43477480 | 0x6e4000 |
|
|
| 79 | blk.6.attn_q.bias | 0x43b5b480 | 0x3800 |
|
|
| 80 | blk.6.attn_q.weight | 0x43b5ec80 | 0x4b0800 |
|
|
| 81 | blk.6.attn_v.bias | 0x4400f480 | 0x800 |
|
|
| 82 | blk.6.attn_v.weight | 0x4400fc80 | 0xc0800 |
|
|
| 83 | blk.6.ffn_down.weight | 0x440d0480 | 0x246c000 |
|
|
| 84 | blk.6.ffn_gate.weight | 0x4653c480 | 0x18c9800 |
|
|
| 85 | blk.6.ffn_norm.weight | 0x47e05c80 | 0x3800 |
|
|
| 86 | blk.6.ffn_up.weight | 0x47e09480 | 0x18c9800 |
|
|
| 87 | blk.7.attn_k.bias | 0x496d2c80 | 0x800 |
|
|
| 88 | blk.7.attn_k.weight | 0x496d3480 | 0xab800 |
|
|
| 89 | blk.7.attn_norm.weight | 0x4977ec80 | 0x3800 |
|
|
| 90 | blk.7.attn_output.weight | 0x49782480 | 0x6e4000 |
|
|
| 91 | blk.7.attn_q.bias | 0x49e66480 | 0x3800 |
|
|
| 92 | blk.7.attn_q.weight | 0x49e69c80 | 0x4b0800 |
|
|
| 93 | blk.7.attn_v.bias | 0x4a31a480 | 0x800 |
|
|
| 94 | blk.7.attn_v.weight | 0x4a31ac80 | 0xc0800 |
|
|
| 95 | blk.7.ffn_down.weight | 0x4a3db480 | 0x246c000 |
|
|
| 96 | blk.7.ffn_gate.weight | 0x4c847480 | 0x18c9800 |
|
|
| 97 | blk.7.ffn_norm.weight | 0x4e110c80 | 0x3800 |
|
|
| 98 | blk.7.ffn_up.weight | 0x4e114480 | 0x18c9800 |
|
|
| 99 | blk.8.attn_k.bias | 0x4f9ddc80 | 0x800 |
|
|
| 100 | blk.8.attn_k.weight | 0x4f9de480 | 0xab800 |
|
|
| 101 | blk.8.attn_norm.weight | 0x4fa89c80 | 0x3800 |
|
|
| 102 | blk.8.attn_output.weight | 0x4fa8d480 | 0x6e4000 |
|
|
| 103 | blk.8.attn_q.bias | 0x50171480 | 0x3800 |
|
|
| 104 | blk.8.attn_q.weight | 0x50174c80 | 0x4b0800 |
|
|
| 105 | blk.8.attn_v.bias | 0x50625480 | 0x800 |
|
|
| 106 | blk.8.attn_v.weight | 0x50625c80 | 0xc0800 |
|
|
| 107 | blk.8.ffn_down.weight | 0x506e6480 | 0x246c000 |
|
|
| 108 | blk.8.ffn_gate.weight | 0x52b52480 | 0x18c9800 |
|
|
| 109 | blk.8.ffn_norm.weight | 0x5441bc80 | 0x3800 |
|
|
| 110 | blk.8.ffn_up.weight | 0x5441f480 | 0x18c9800 |
|
|
| 111 | blk.9.attn_k.bias | 0x55ce8c80 | 0x800 |
|
|
| 112 | blk.9.attn_k.weight | 0x55ce9480 | 0xab800 |
|
|
| 113 | blk.9.attn_norm.weight | 0x55d94c80 | 0x3800 |
|
|
| 114 | blk.9.attn_output.weight | 0x55d98480 | 0x6e4000 |
|
|
| 115 | blk.9.attn_q.bias | 0x5647c480 | 0x3800 |
|
|
| 116 | blk.9.attn_q.weight | 0x5647fc80 | 0x4b0800 |
|
|
| 117 | blk.9.attn_v.bias | 0x56930480 | 0x800 |
|
|
| 118 | blk.9.attn_v.weight | 0x56930c80 | 0xc0800 |
|
|
| 119 | blk.9.ffn_down.weight | 0x569f1480 | 0x246c000 |
|
|
| 120 | blk.9.ffn_gate.weight | 0x58e5d480 | 0x1bd2800 |
|
|
| 121 | blk.9.ffn_norm.weight | 0x5aa2fc80 | 0x3800 |
|
|
| 122 | blk.9.ffn_up.weight | 0x5aa33480 | 0x1bd2800 |
|
|
| 123 | blk.10.attn_k.bias | 0x5c605c80 | 0x800 |
|
|
| 124 | blk.10.attn_k.weight | 0x5c606480 | 0xab800 |
|
|
| 125 | blk.10.attn_norm.weight | 0x5c6b1c80 | 0x3800 |
|
|
| 126 | blk.10.attn_output.weight | 0x5c6b5480 | 0x6e4000 |
|
|
| 127 | blk.10.attn_q.bias | 0x5cd99480 | 0x3800 |
|
|
| 128 | blk.10.attn_q.weight | 0x5cd9cc80 | 0x4b0800 |
|
|
| 129 | blk.10.attn_v.bias | 0x5d24d480 | 0x800 |
|
|
| 130 | blk.10.attn_v.weight | 0x5d24dc80 | 0xc0800 |
|
|
| 131 | blk.10.ffn_down.weight | 0x5d30e480 | 0x246c000 |
|
|
| 132 | blk.10.ffn_gate.weight | 0x5f77a480 | 0x18c9800 |
|
|
| 133 | blk.10.ffn_norm.weight | 0x61043c80 | 0x3800 |
|
|
| 134 | blk.10.ffn_up.weight | 0x61047480 | 0x18c9800 |
|
|
| 135 | blk.11.attn_k.bias | 0x62910c80 | 0x800 |
|
|
| 136 | blk.11.attn_k.weight | 0x62911480 | 0xab800 |
|
|
| 137 | blk.11.attn_norm.weight | 0x629bcc80 | 0x3800 |
|
|
| 138 | blk.11.attn_output.weight | 0x629c0480 | 0x6e4000 |
|
|
| 139 | blk.11.attn_q.bias | 0x630a4480 | 0x3800 |
|
|
| 140 | blk.11.attn_q.weight | 0x630a7c80 | 0x4b0800 |
|
|
| 141 | blk.11.attn_v.bias | 0x63558480 | 0x800 |
|
|
| 142 | blk.11.attn_v.weight | 0x63558c80 | 0xc0800 |
|
|
| 143 | blk.11.ffn_down.weight | 0x63619480 | 0x246c000 |
|
|
| 144 | blk.11.ffn_gate.weight | 0x65a85480 | 0x18c9800 |
|
|
| 145 | blk.11.ffn_norm.weight | 0x6734ec80 | 0x3800 |
|
|
| 146 | blk.11.ffn_up.weight | 0x67352480 | 0x18c9800 |
|
|
| 147 | blk.12.attn_k.bias | 0x68c1bc80 | 0x800 |
|
|
| 148 | blk.12.attn_k.weight | 0x68c1c480 | 0xab800 |
|
|
| 149 | blk.12.attn_norm.weight | 0x68cc7c80 | 0x3800 |
|
|
| 150 | blk.12.attn_output.weight | 0x68ccb480 | 0x6e4000 |
|
|
| 151 | blk.12.attn_q.bias | 0x693af480 | 0x3800 |
|
|
| 152 | blk.12.attn_q.weight | 0x693b2c80 | 0x4b0800 |
|
|
| 153 | blk.12.attn_v.bias | 0x69863480 | 0x800 |
|
|
| 154 | blk.12.attn_v.weight | 0x69863c80 | 0xc0800 |
|
|
| 155 | blk.12.ffn_down.weight | 0x69924480 | 0x246c000 |
|
|
| 156 | blk.12.ffn_gate.weight | 0x6bd90480 | 0x18c9800 |
|
|
| 157 | blk.12.ffn_norm.weight | 0x6d659c80 | 0x3800 |
|
|
| 158 | blk.12.ffn_up.weight | 0x6d65d480 | 0x18c9800 |
|
|
| 159 | blk.13.attn_k.bias | 0x6ef26c80 | 0x800 |
|
|
| 160 | blk.13.attn_k.weight | 0x6ef27480 | 0xab800 |
|
|
| 161 | blk.13.attn_norm.weight | 0x6efd2c80 | 0x3800 |
|
|
| 162 | blk.13.attn_output.weight | 0x6efd6480 | 0x6e4000 |
|
|
| 163 | blk.13.attn_q.bias | 0x6f6ba480 | 0x3800 |
|
|
| 164 | blk.13.attn_q.weight | 0x6f6bdc80 | 0x4b0800 |
|
|
| 165 | blk.13.attn_v.bias | 0x6fb6e480 | 0x800 |
|
|
| 166 | blk.13.attn_v.weight | 0x6fb6ec80 | 0xc0800 |
|
|
| 167 | blk.13.ffn_down.weight | 0x6fc2f480 | 0x246c000 |
|
|
| 168 | blk.13.ffn_gate.weight | 0x7209b480 | 0x18c9800 |
|
|
| 169 | blk.13.ffn_norm.weight | 0x73964c80 | 0x3800 |
|
|
| 170 | blk.13.ffn_up.weight | 0x73968480 | 0x18c9800 |
|
|
| 171 | blk.14.attn_k.bias | 0x75231c80 | 0x800 |
|
|
| 172 | blk.14.attn_k.weight | 0x75232480 | 0xc0800 |
|
|
| 173 | blk.14.attn_norm.weight | 0x752f2c80 | 0x3800 |
|
|
| 174 | blk.14.attn_output.weight | 0x752f6480 | 0x6e4000 |
|
|
| 175 | blk.14.attn_q.bias | 0x759da480 | 0x3800 |
|
|
| 176 | blk.14.attn_q.weight | 0x759ddc80 | 0x543800 |
|
|
| 177 | blk.14.attn_v.bias | 0x75f21480 | 0x800 |
|
|
| 178 | blk.14.attn_v.weight | 0x75f21c80 | 0xfc000 |
|
|
| 179 | blk.14.ffn_down.weight | 0x7601dc80 | 0x246c000 |
|
|
| 180 | blk.14.ffn_gate.weight | 0x78489c80 | 0x18c9800 |
|
|
| 181 | blk.14.ffn_norm.weight | 0x79d53480 | 0x3800 |
|
|
| 182 | blk.14.ffn_up.weight | 0x79d56c80 | 0x18c9800 |
|
|
| 183 | blk.15.attn_k.bias | 0x7b620480 | 0x800 |
|
|
| 184 | blk.15.attn_k.weight | 0x7b620c80 | 0xc0800 |
|
|
| 185 | blk.15.attn_norm.weight | 0x7b6e1480 | 0x3800 |
|
|
| 186 | blk.15.attn_output.weight | 0x7b6e4c80 | 0x6e4000 |
|
|
| 187 | blk.15.attn_q.bias | 0x7bdc8c80 | 0x3800 |
|
|
| 188 | blk.15.attn_q.weight | 0x7bdcc480 | 0x543800 |
|
|
| 189 | blk.15.attn_v.bias | 0x7c30fc80 | 0x800 |
|
|
| 190 | blk.15.attn_v.weight | 0x7c310480 | 0xfc000 |
|
|
| 191 | blk.15.ffn_down.weight | 0x7c40c480 | 0x246c000 |
|
|
| 192 | blk.15.ffn_gate.weight | 0x7e878480 | 0x18c9800 |
|
|
| 193 | blk.15.ffn_norm.weight | 0x80141c80 | 0x3800 |
|
|
| 194 | blk.15.ffn_up.weight | 0x80145480 | 0x18c9800 |
|
|
| 195 | blk.16.attn_k.bias | 0x81a0ec80 | 0x800 |
|
|
| 196 | blk.16.attn_k.weight | 0x81a0f480 | 0xc0800 |
|
|
| 197 | blk.16.attn_norm.weight | 0x81acfc80 | 0x3800 |
|
|
| 198 | blk.16.attn_output.weight | 0x81ad3480 | 0x6e4000 |
|
|
| 199 | blk.16.attn_q.bias | 0x821b7480 | 0x3800 |
|
|
| 200 | blk.16.attn_q.weight | 0x821bac80 | 0x543800 |
|
|
| 201 | blk.16.attn_v.bias | 0x826fe480 | 0x800 |
|
|
| 202 | blk.16.attn_v.weight | 0x826fec80 | 0xfc000 |
|
|
| 203 | blk.16.ffn_down.weight | 0x827fac80 | 0x246c000 |
|
|
| 204 | blk.16.ffn_gate.weight | 0x84c66c80 | 0x18c9800 |
|
|
| 205 | blk.16.ffn_norm.weight | 0x86530480 | 0x3800 |
|
|
| 206 | blk.16.ffn_up.weight | 0x86533c80 | 0x18c9800 |
|
|
| 207 | blk.17.attn_k.bias | 0x87dfd480 | 0x800 |
|
|
| 208 | blk.17.attn_k.weight | 0x87dfdc80 | 0xc0800 |
|
|
| 209 | blk.17.attn_norm.weight | 0x87ebe480 | 0x3800 |
|
|
| 210 | blk.17.attn_output.weight | 0x87ec1c80 | 0x6e4000 |
|
|
| 211 | blk.17.attn_q.bias | 0x885a5c80 | 0x3800 |
|
|
| 212 | blk.17.attn_q.weight | 0x885a9480 | 0x543800 |
|
|
| 213 | blk.17.attn_v.bias | 0x88aecc80 | 0x800 |
|
|
| 214 | blk.17.attn_v.weight | 0x88aed480 | 0xfc000 |
|
|
| 215 | blk.17.ffn_down.weight | 0x88be9480 | 0x246c000 |
|
|
| 216 | blk.17.ffn_gate.weight | 0x8b055480 | 0x18c9800 |
|
|
| 217 | blk.17.ffn_norm.weight | 0x8c91ec80 | 0x3800 |
|
|
| 218 | blk.17.ffn_up.weight | 0x8c922480 | 0x18c9800 |
|
|
| 219 | blk.18.attn_k.bias | 0x8e1ebc80 | 0x800 |
|
|
| 220 | blk.18.attn_k.weight | 0x8e1ec480 | 0xc0800 |
|
|
| 221 | blk.18.attn_norm.weight | 0x8e2acc80 | 0x3800 |
|
|
| 222 | blk.18.attn_output.weight | 0x8e2b0480 | 0x6e4000 |
|
|
| 223 | blk.18.attn_q.bias | 0x8e994480 | 0x3800 |
|
|
| 224 | blk.18.attn_q.weight | 0x8e997c80 | 0x543800 |
|
|
| 225 | blk.18.attn_v.bias | 0x8eedb480 | 0x800 |
|
|
| 226 | blk.18.attn_v.weight | 0x8eedbc80 | 0xfc000 |
|
|
| 227 | blk.18.ffn_down.weight | 0x8efd7c80 | 0x246c000 |
|
|
| 228 | blk.18.ffn_gate.weight | 0x91443c80 | 0x18c9800 |
|
|
| 229 | blk.18.ffn_norm.weight | 0x92d0d480 | 0x3800 |
|
|
| 230 | blk.18.ffn_up.weight | 0x92d10c80 | 0x18c9800 |
|
|
| 231 | blk.19.attn_k.bias | 0x945da480 | 0x800 |
|
|
| 232 | blk.19.attn_k.weight | 0x945dac80 | 0xc0800 |
|
|
| 233 | blk.19.attn_norm.weight | 0x9469b480 | 0x3800 |
|
|
| 234 | blk.19.attn_output.weight | 0x9469ec80 | 0x6e4000 |
|
|
| 235 | blk.19.attn_q.bias | 0x94d82c80 | 0x3800 |
|
|
| 236 | blk.19.attn_q.weight | 0x94d86480 | 0x543800 |
|
|
| 237 | blk.19.attn_v.bias | 0x952c9c80 | 0x800 |
|
|
| 238 | blk.19.attn_v.weight | 0x952ca480 | 0xfc000 |
|
|
| 239 | blk.19.ffn_down.weight | 0x953c6480 | 0x246c000 |
|
|
| 240 | blk.19.ffn_gate.weight | 0x97832480 | 0x18c9800 |
|
|
| 241 | blk.19.ffn_norm.weight | 0x990fbc80 | 0x3800 |
|
|
| 242 | blk.19.ffn_up.weight | 0x990ff480 | 0x18c9800 |
|
|
| 243 | blk.20.attn_k.bias | 0x9a9c8c80 | 0x800 |
|
|
| 244 | blk.20.attn_k.weight | 0x9a9c9480 | 0xc0800 |
|
|
| 245 | blk.20.attn_norm.weight | 0x9aa89c80 | 0x3800 |
|
|
| 246 | blk.20.attn_output.weight | 0x9aa8d480 | 0x6e4000 |
|
|
| 247 | blk.20.attn_q.bias | 0x9b171480 | 0x3800 |
|
|
| 248 | blk.20.attn_q.weight | 0x9b174c80 | 0x543800 |
|
|
| 249 | blk.20.attn_v.bias | 0x9b6b8480 | 0x800 |
|
|
| 250 | blk.20.attn_v.weight | 0x9b6b8c80 | 0xfc000 |
|
|
| 251 | blk.20.ffn_down.weight | 0x9b7b4c80 | 0x246c000 |
|
|
| 252 | blk.20.ffn_gate.weight | 0x9dc20c80 | 0x1bd2800 |
|
|
| 253 | blk.20.ffn_norm.weight | 0x9f7f3480 | 0x3800 |
|
|
| 254 | blk.20.ffn_up.weight | 0x9f7f6c80 | 0x1bd2800 |
|
|
| 255 | blk.21.attn_k.bias | 0xa13c9480 | 0x800 |
|
|
| 256 | blk.21.attn_k.weight | 0xa13c9c80 | 0xc0800 |
|
|
| 257 | blk.21.attn_norm.weight | 0xa148a480 | 0x3800 |
|
|
| 258 | blk.21.attn_output.weight | 0xa148dc80 | 0x6e4000 |
|
|
| 259 | blk.21.attn_q.bias | 0xa1b71c80 | 0x3800 |
|
|
| 260 | blk.21.attn_q.weight | 0xa1b75480 | 0x543800 |
|
|
| 261 | blk.21.attn_v.bias | 0xa20b8c80 | 0x800 |
|
|
| 262 | blk.21.attn_v.weight | 0xa20b9480 | 0xfc000 |
|
|
| 263 | blk.21.ffn_down.weight | 0xa21b5480 | 0x246c000 |
|
|
| 264 | blk.21.ffn_gate.weight | 0xa4621480 | 0x1bd2800 |
|
|
| 265 | blk.21.ffn_norm.weight | 0xa61f3c80 | 0x3800 |
|
|
| 266 | blk.21.ffn_up.weight | 0xa61f7480 | 0x1bd2800 |
|
|
| 267 | blk.22.attn_k.bias | 0xa7dc9c80 | 0x800 |
|
|
| 268 | blk.22.attn_k.weight | 0xa7dca480 | 0xc0800 |
|
|
| 269 | blk.22.attn_norm.weight | 0xa7e8ac80 | 0x3800 |
|
|
| 270 | blk.22.attn_output.weight | 0xa7e8e480 | 0x6e4000 |
|
|
| 271 | blk.22.attn_q.bias | 0xa8572480 | 0x3800 |
|
|
| 272 | blk.22.attn_q.weight | 0xa8575c80 | 0x543800 |
|
|
| 273 | blk.22.attn_v.bias | 0xa8ab9480 | 0x800 |
|
|
| 274 | blk.22.attn_v.weight | 0xa8ab9c80 | 0xfc000 |
|
|
| 275 | blk.22.ffn_down.weight | 0xa8bb5c80 | 0x246c000 |
|
|
| 276 | blk.22.ffn_gate.weight | 0xab021c80 | 0x1bd2800 |
|
|
| 277 | blk.22.ffn_norm.weight | 0xacbf4480 | 0x3800 |
|
|
| 278 | blk.22.ffn_up.weight | 0xacbf7c80 | 0x1bd2800 |
|
|
| 279 | blk.23.attn_k.bias | 0xae7ca480 | 0x800 |
|
|
| 280 | blk.23.attn_k.weight | 0xae7cac80 | 0xc0800 |
|
|
| 281 | blk.23.attn_norm.weight | 0xae88b480 | 0x3800 |
|
|
| 282 | blk.23.attn_output.weight | 0xae88ec80 | 0x6e4000 |
|
|
| 283 | blk.23.attn_q.bias | 0xaef72c80 | 0x3800 |
|
|
| 284 | blk.23.attn_q.weight | 0xaef76480 | 0x543800 |
|
|
| 285 | blk.23.attn_v.bias | 0xaf4b9c80 | 0x800 |
|
|
| 286 | blk.23.attn_v.weight | 0xaf4ba480 | 0xfc000 |
|
|
| 287 | blk.23.ffn_down.weight | 0xaf5b6480 | 0x246c000 |
|
|
| 288 | blk.23.ffn_gate.weight | 0xb1a22480 | 0x1bd2800 |
|
|
| 289 | blk.23.ffn_norm.weight | 0xb35f4c80 | 0x3800 |
|
|
| 290 | blk.23.ffn_up.weight | 0xb35f8480 | 0x1bd2800 |
|
|
| 291 | blk.24.attn_k.bias | 0xb51cac80 | 0x800 |
|
|
| 292 | blk.24.attn_k.weight | 0xb51cb480 | 0xc0800 |
|
|
| 293 | blk.24.attn_norm.weight | 0xb528bc80 | 0x3800 |
|
|
| 294 | blk.24.attn_output.weight | 0xb528f480 | 0x6e4000 |
|
|
| 295 | blk.24.attn_q.bias | 0xb5973480 | 0x3800 |
|
|
| 296 | blk.24.attn_q.weight | 0xb5976c80 | 0x543800 |
|
|
| 297 | blk.24.attn_v.bias | 0xb5eba480 | 0x800 |
|
|
| 298 | blk.24.attn_v.weight | 0xb5ebac80 | 0xfc000 |
|
|
| 299 | blk.24.ffn_down.weight | 0xb5fb6c80 | 0x246c000 |
|
|
| 300 | blk.24.ffn_gate.weight | 0xb8422c80 | 0x1bd2800 |
|
|
| 301 | blk.24.ffn_norm.weight | 0xb9ff5480 | 0x3800 |
|
|
| 302 | blk.24.ffn_up.weight | 0xb9ff8c80 | 0x1bd2800 |
|
|
| 303 | blk.25.attn_k.bias | 0xbbbcb480 | 0x800 |
|
|
| 304 | blk.25.attn_k.weight | 0xbbbcbc80 | 0xc0800 |
|
|
| 305 | blk.25.attn_norm.weight | 0xbbc8c480 | 0x3800 |
|
|
| 306 | blk.25.attn_output.weight | 0xbbc8fc80 | 0x6e4000 |
|
|
| 307 | blk.25.attn_q.bias | 0xbc373c80 | 0x3800 |
|
|
| 308 | blk.25.attn_q.weight | 0xbc377480 | 0x543800 |
|
|
| 309 | blk.25.attn_v.bias | 0xbc8bac80 | 0x800 |
|
|
| 310 | blk.25.attn_v.weight | 0xbc8bb480 | 0xfc000 |
|
|
| 311 | blk.25.ffn_down.weight | 0xbc9b7480 | 0x246c000 |
|
|
| 312 | blk.25.ffn_gate.weight | 0xbee23480 | 0x1bd2800 |
|
|
| 313 | blk.25.ffn_norm.weight | 0xc09f5c80 | 0x3800 |
|
|
| 314 | blk.25.ffn_up.weight | 0xc09f9480 | 0x1bd2800 |
|
|
| 315 | blk.26.attn_k.bias | 0xc25cbc80 | 0x800 |
|
|
| 316 | blk.26.attn_k.weight | 0xc25cc480 | 0xc0800 |
|
|
| 317 | blk.26.attn_norm.weight | 0xc268cc80 | 0x3800 |
|
|
| 318 | blk.26.attn_output.weight | 0xc2690480 | 0x6e4000 |
|
|
| 319 | blk.26.attn_q.bias | 0xc2d74480 | 0x3800 |
|
|
| 320 | blk.26.attn_q.weight | 0xc2d77c80 | 0x543800 |
|
|
| 321 | blk.26.attn_v.bias | 0xc32bb480 | 0x800 |
|
|
| 322 | blk.26.attn_v.weight | 0xc32bbc80 | 0xfc000 |
|
|
| 323 | blk.26.ffn_down.weight | 0xc33b7c80 | 0x246c000 |
|
|
| 324 | blk.26.ffn_gate.weight | 0xc5823c80 | 0x1bd2800 |
|
|
| 325 | blk.26.ffn_norm.weight | 0xc73f6480 | 0x3800 |
|
|
| 326 | blk.26.ffn_up.weight | 0xc73f9c80 | 0x1bd2800 |
|
|
| 327 | blk.27.attn_k.bias | 0xc8fcc480 | 0x800 |
|
|
| 328 | blk.27.attn_k.weight | 0xc8fccc80 | 0xc0800 |
|
|
| 329 | blk.27.attn_norm.weight | 0xc908d480 | 0x3800 |
|
|
| 330 | blk.27.attn_output.weight | 0xc9090c80 | 0x6e4000 |
|
|
| 331 | blk.27.attn_q.bias | 0xc9774c80 | 0x3800 |
|
|
| 332 | blk.27.attn_q.weight | 0xc9778480 | 0x543800 |
|
|
| 333 | blk.27.attn_v.bias | 0xc9cbbc80 | 0x800 |
|
|
| 334 | blk.27.attn_v.weight | 0xc9cbc480 | 0xfc000 |
|
|
| 335 | blk.27.ffn_down.weight | 0xc9db8480 | 0x246c000 |
|
|
| 336 | blk.27.ffn_gate.weight | 0xcc224480 | 0x1bd2800 |
|
|
| 337 | blk.27.ffn_norm.weight | 0xcddf6c80 | 0x3800 |
|
|
| 338 | blk.27.ffn_up.weight | 0xcddfa480 | 0x1bd2800 |
|
|
|
|
### <a name="base">Base Tensor Group : ~1B Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:------|
|
|
| 0 | output.weight | Output (W) | (~544M) 543567360 | 3584 x 151665 x 1 x 1 | IQ3_S |
|
|
| 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 2 | token_embd.weight | Token Embedding (W) | (~544M) 543567360 | 3584 x 151665 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in base: ( ~1B) 1087138304
|
|
- Percentage of total elements: 14.28%
|
|
|
|
|
|
### <a name="blk_0">Block 0 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 3 | blk.0.attn_k.bias | Block 0 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 7 | blk.0.attn_q.bias | Block 0 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 8 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 9 | blk.0.attn_v.bias | Block 0 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 10 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 11 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 12 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 13 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 14 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.0: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_1">Block 1 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 15 | blk.1.attn_k.bias | Block 1 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 16 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 17 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 18 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 19 | blk.1.attn_q.bias | Block 1 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 20 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 21 | blk.1.attn_v.bias | Block 1 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 22 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 23 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 24 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 25 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 26 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.1: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_2">Block 2 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 27 | blk.2.attn_k.bias | Block 2 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 28 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 29 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 30 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 31 | blk.2.attn_q.bias | Block 2 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 32 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 33 | blk.2.attn_v.bias | Block 2 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 34 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 35 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 36 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 37 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 38 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.2: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_3">Block 3 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 39 | blk.3.attn_k.bias | Block 3 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 40 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 41 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 42 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 43 | blk.3.attn_q.bias | Block 3 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 44 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 45 | blk.3.attn_v.bias | Block 3 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 46 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 47 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 48 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 49 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 50 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.3: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_4">Block 4 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 51 | blk.4.attn_k.bias | Block 4 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 52 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 53 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 54 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 55 | blk.4.attn_q.bias | Block 4 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 56 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 57 | blk.4.attn_v.bias | Block 4 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 58 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 59 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 60 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 61 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 62 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.4: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_5">Block 5 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 63 | blk.5.attn_k.bias | Block 5 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 64 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 65 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 66 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 67 | blk.5.attn_q.bias | Block 5 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 68 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 69 | blk.5.attn_v.bias | Block 5 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 70 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 71 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 72 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 73 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 74 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.5: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_6">Block 6 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 75 | blk.6.attn_k.bias | Block 6 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 76 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 77 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 78 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 79 | blk.6.attn_q.bias | Block 6 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 80 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 81 | blk.6.attn_v.bias | Block 6 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 82 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 83 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 84 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 85 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 86 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.6: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_7">Block 7 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 87 | blk.7.attn_k.bias | Block 7 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 88 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 89 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 90 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 91 | blk.7.attn_q.bias | Block 7 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 92 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 93 | blk.7.attn_v.bias | Block 7 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 94 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 95 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 96 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 97 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 98 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.7: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_8">Block 8 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 99 | blk.8.attn_k.bias | Block 8 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 100 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 101 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 102 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 103 | blk.8.attn_q.bias | Block 8 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 104 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 105 | blk.8.attn_v.bias | Block 8 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 106 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 107 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 108 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 109 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 110 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.8: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_9">Block 9 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 111 | blk.9.attn_k.bias | Block 9 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 112 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 113 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 114 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 115 | blk.9.attn_q.bias | Block 9 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 116 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 117 | blk.9.attn_v.bias | Block 9 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 118 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 119 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 120 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 121 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 122 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.9: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_10">Block 10 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 123 | blk.10.attn_k.bias | Block 10 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 124 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 125 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 126 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 127 | blk.10.attn_q.bias | Block 10 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 128 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 129 | blk.10.attn_v.bias | Block 10 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 130 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 131 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 132 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 133 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 134 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.10: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_11">Block 11 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 135 | blk.11.attn_k.bias | Block 11 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 136 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 137 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 138 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 139 | blk.11.attn_q.bias | Block 11 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 140 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 141 | blk.11.attn_v.bias | Block 11 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 142 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 143 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 144 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 145 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 146 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.11: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_12">Block 12 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 147 | blk.12.attn_k.bias | Block 12 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 148 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 149 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 150 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 151 | blk.12.attn_q.bias | Block 12 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 152 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 153 | blk.12.attn_v.bias | Block 12 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 154 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 155 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 156 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 157 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 158 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.12: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_13">Block 13 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 159 | blk.13.attn_k.bias | Block 13 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 160 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_XXS |
|
|
| 161 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 162 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 163 | blk.13.attn_q.bias | Block 13 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 164 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_XXS |
|
|
| 165 | blk.13.attn_v.bias | Block 13 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 166 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 167 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 168 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 169 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 170 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.13: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_14">Block 14 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 171 | blk.14.attn_k.bias | Block 14 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 172 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 173 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 174 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 175 | blk.14.attn_q.bias | Block 14 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 176 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 177 | blk.14.attn_v.bias | Block 14 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 178 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 179 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 180 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 181 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 182 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.14: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_15">Block 15 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 183 | blk.15.attn_k.bias | Block 15 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 184 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 185 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 186 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 187 | blk.15.attn_q.bias | Block 15 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 188 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 189 | blk.15.attn_v.bias | Block 15 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 190 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 191 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 192 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 193 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 194 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.15: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_16">Block 16 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 195 | blk.16.attn_k.bias | Block 16 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 196 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 197 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 198 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 199 | blk.16.attn_q.bias | Block 16 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 200 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 201 | blk.16.attn_v.bias | Block 16 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 202 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 203 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 204 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 205 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 206 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.16: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_17">Block 17 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 207 | blk.17.attn_k.bias | Block 17 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 208 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 209 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 210 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 211 | blk.17.attn_q.bias | Block 17 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 212 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 213 | blk.17.attn_v.bias | Block 17 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 214 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 215 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 216 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 217 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 218 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.17: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_18">Block 18 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 219 | blk.18.attn_k.bias | Block 18 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 220 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 221 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 222 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 223 | blk.18.attn_q.bias | Block 18 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 224 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 225 | blk.18.attn_v.bias | Block 18 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 226 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 227 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 228 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 229 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 230 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.18: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_19">Block 19 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
|
|
| 231 | blk.19.attn_k.bias | Block 19 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 232 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 233 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 234 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 235 | blk.19.attn_q.bias | Block 19 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 236 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 237 | blk.19.attn_v.bias | Block 19 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 238 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 239 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 240 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
| 241 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 242 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_XXS |
|
|
|
|
- Total elements in blk.19: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_20">Block 20 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
|
|
| 243 | blk.20.attn_k.bias | Block 20 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 244 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 245 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 246 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 247 | blk.20.attn_q.bias | Block 20 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 248 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 249 | blk.20.attn_v.bias | Block 20 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 250 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 251 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 252 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 253 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 254 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.20: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_21">Block 21 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
|
|
| 255 | blk.21.attn_k.bias | Block 21 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 256 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 257 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 258 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 259 | blk.21.attn_q.bias | Block 21 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 260 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 261 | blk.21.attn_v.bias | Block 21 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 262 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 263 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 264 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 265 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 266 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.21: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_22">Block 22 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
|
|
| 267 | blk.22.attn_k.bias | Block 22 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 268 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 269 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 270 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 271 | blk.22.attn_q.bias | Block 22 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 272 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 273 | blk.22.attn_v.bias | Block 22 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 274 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 275 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 276 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 277 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 278 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.22: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_23">Block 23 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
|
|
| 279 | blk.23.attn_k.bias | Block 23 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 280 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 281 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 282 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 283 | blk.23.attn_q.bias | Block 23 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 284 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 285 | blk.23.attn_v.bias | Block 23 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 286 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 287 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 288 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 289 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 290 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.23: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_24">Block 24 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
|
|
| 291 | blk.24.attn_k.bias | Block 24 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 292 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 293 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 294 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 295 | blk.24.attn_q.bias | Block 24 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 296 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 297 | blk.24.attn_v.bias | Block 24 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 298 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 299 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 300 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 301 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 302 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.24: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_25">Block 25 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
|
|
| 303 | blk.25.attn_k.bias | Block 25 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 304 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 305 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 306 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 307 | blk.25.attn_q.bias | Block 25 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 308 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 309 | blk.25.attn_v.bias | Block 25 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 310 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 311 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 312 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 313 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 314 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.25: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_26">Block 26 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
|
|
| 315 | blk.26.attn_k.bias | Block 26 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 316 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 317 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 318 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 319 | blk.26.attn_q.bias | Block 26 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 320 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 321 | blk.26.attn_v.bias | Block 26 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 322 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 323 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 324 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 325 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 326 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.26: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|
|
|
|
|
|
### <a name="blk_27">Block 27 Tensor Group : ~233M Elements</a>
|
|
|
|
| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
|
|
|-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
|
|
| 327 | blk.27.attn_k.bias | Block 27 Attention Key (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 328 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ3_S |
|
|
| 329 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 330 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | Q4_K |
|
|
| 331 | blk.27.attn_q.bias | Block 27 Attention Query (B) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 332 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~13M) 12845056 | 3584 x 3584 x 1 x 1 | IQ3_S |
|
|
| 333 | blk.27.attn_v.bias | Block 27 Attention Value (B) | ( 512) 512 | 512 x 1 x 1 x 1 | F32 |
|
|
| 334 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~2M) 1835008 | 3584 x 512 x 1 x 1 | IQ4_NL |
|
|
| 335 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~68M) 67895296 | 18944 x 3584 x 1 x 1 | IQ4_NL |
|
|
| 336 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
| 337 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 3584 | 3584 x 1 x 1 x 1 | F32 |
|
|
| 338 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~68M) 67895296 | 3584 x 18944 x 1 x 1 | IQ3_S |
|
|
|
|
- Total elements in blk.27: (~233M) 233057792
|
|
- Percentage of total elements: 3.06%
|