From 5f0ed08c7743f24519bca59b4d91077916c7e379 Mon Sep 17 00:00:00 2001 From: Aleksey Korshuk Date: Wed, 5 Apr 2023 03:34:44 +0000 Subject: [PATCH] Upload LlamaForCausalLM --- config.json | 23 +++ generation_config.json | 7 + pytorch_model-00001-of-00014.bin | 3 + pytorch_model-00002-of-00014.bin | 3 + pytorch_model-00003-of-00014.bin | 3 + pytorch_model-00004-of-00014.bin | 3 + pytorch_model-00005-of-00014.bin | 3 + pytorch_model-00006-of-00014.bin | 3 + pytorch_model-00007-of-00014.bin | 3 + pytorch_model-00008-of-00014.bin | 3 + pytorch_model-00009-of-00014.bin | 3 + pytorch_model-00010-of-00014.bin | 3 + pytorch_model-00011-of-00014.bin | 3 + pytorch_model-00012-of-00014.bin | 3 + pytorch_model-00013-of-00014.bin | 3 + pytorch_model-00014-of-00014.bin | 3 + pytorch_model.bin.index.json | 330 +++++++++++++++++++++++++++++++ 17 files changed, 402 insertions(+) create mode 100644 config.json create mode 100644 generation_config.json create mode 100644 pytorch_model-00001-of-00014.bin create mode 100644 pytorch_model-00002-of-00014.bin create mode 100644 pytorch_model-00003-of-00014.bin create mode 100644 pytorch_model-00004-of-00014.bin create mode 100644 pytorch_model-00005-of-00014.bin create mode 100644 pytorch_model-00006-of-00014.bin create mode 100644 pytorch_model-00007-of-00014.bin create mode 100644 pytorch_model-00008-of-00014.bin create mode 100644 pytorch_model-00009-of-00014.bin create mode 100644 pytorch_model-00010-of-00014.bin create mode 100644 pytorch_model-00011-of-00014.bin create mode 100644 pytorch_model-00012-of-00014.bin create mode 100644 pytorch_model-00013-of-00014.bin create mode 100644 pytorch_model-00014-of-00014.bin create mode 100644 pytorch_model.bin.index.json diff --git a/config.json b/config.json new file mode 100644 index 0000000..fa01c13 --- /dev/null +++ b/config.json @@ -0,0 +1,23 @@ +{ + "_name_or_path": "./", + "architectures": [ + "LlamaForCausalLM" + ], + "bos_token_id": 0, + "eos_token_id": 1, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 11008, + "max_sequence_length": 2048, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "pad_token_id": -1, + "rms_norm_eps": 1e-06, + "tie_word_embeddings": false, + "torch_dtype": "float16", + "transformers_version": "4.28.0.dev0", + "use_cache": true, + "vocab_size": 32001 +} diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000..128b489 --- /dev/null +++ b/generation_config.json @@ -0,0 +1,7 @@ +{ + "_from_model_config": true, + "bos_token_id": 0, + "eos_token_id": 1, + "pad_token_id": 0, + "transformers_version": "4.28.0.dev0" +} diff --git a/pytorch_model-00001-of-00014.bin b/pytorch_model-00001-of-00014.bin new file mode 100644 index 0000000..efef8b9 --- /dev/null +++ b/pytorch_model-00001-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77d715580dfc4bf1fe0ae1498f27e7192ad3ff0df924a69c4e80d72b51b146ba +size 981498195 diff --git a/pytorch_model-00002-of-00014.bin b/pytorch_model-00002-of-00014.bin new file mode 100644 index 0000000..8978b4c --- /dev/null +++ b/pytorch_model-00002-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:583ac5dba64145de72674674eef88cb1319441bd0b870ebc80ced4dede6faa80 +size 966844881 diff --git a/pytorch_model-00003-of-00014.bin b/pytorch_model-00003-of-00014.bin new file mode 100644 index 0000000..0da2d78 --- /dev/null +++ b/pytorch_model-00003-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:74152c3c79834987e4db5ca3725c322cdbb1bafdc09031f1fca74be02b185dfa +size 966828351 diff --git a/pytorch_model-00004-of-00014.bin b/pytorch_model-00004-of-00014.bin new file mode 100644 index 0000000..841be36 --- /dev/null +++ b/pytorch_model-00004-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc6f5d634872863709cd26c36ace46d176f944c9e54b2e5f7714fe4031631fae +size 989913279 diff --git a/pytorch_model-00005-of-00014.bin b/pytorch_model-00005-of-00014.bin new file mode 100644 index 0000000..f11c0a8 --- /dev/null +++ b/pytorch_model-00005-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50952fbaf255d762b314e0b020d43b6c98a43852d35ce00ce9c8be8600a14baf +size 943760017 diff --git a/pytorch_model-00006-of-00014.bin b/pytorch_model-00006-of-00014.bin new file mode 100644 index 0000000..893e04a --- /dev/null +++ b/pytorch_model-00006-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:590339279709ca4214c7dfbeb656fe88aa800f6cc6716cf7e3298dfc118d811d +size 989896283 diff --git a/pytorch_model-00007-of-00014.bin b/pytorch_model-00007-of-00014.bin new file mode 100644 index 0000000..0265679 --- /dev/null +++ b/pytorch_model-00007-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f98247078330aa2b31165495937ea497c9d748c7c1e00ef684237a2c1bce11f +size 966844945 diff --git a/pytorch_model-00008-of-00014.bin b/pytorch_model-00008-of-00014.bin new file mode 100644 index 0000000..99b6105 --- /dev/null +++ b/pytorch_model-00008-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a3a411a165123be9633e4c609e5c217e401ff5ae0e85e7f2e45e0929f814f48 +size 966828351 diff --git a/pytorch_model-00009-of-00014.bin b/pytorch_model-00009-of-00014.bin new file mode 100644 index 0000000..2e00e82 --- /dev/null +++ b/pytorch_model-00009-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:837aafee1064964aeaac76672ef4ae0d60b198edf32c2238f158401a2d2854ad +size 989913279 diff --git a/pytorch_model-00010-of-00014.bin b/pytorch_model-00010-of-00014.bin new file mode 100644 index 0000000..b4965ca --- /dev/null +++ b/pytorch_model-00010-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:719e00fc3845ba66234d162f8e96b7f5b52b9a3a813298bb07b1fc0429bab1a7 +size 943760017 diff --git a/pytorch_model-00011-of-00014.bin b/pytorch_model-00011-of-00014.bin new file mode 100644 index 0000000..caf8a9a --- /dev/null +++ b/pytorch_model-00011-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5d54dc19eaae3f7a6af8c02f817f8fe5694fda5e212827038a4873a4db01dac2 +size 989896283 diff --git a/pytorch_model-00012-of-00014.bin b/pytorch_model-00012-of-00014.bin new file mode 100644 index 0000000..9bb5bd0 --- /dev/null +++ b/pytorch_model-00012-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5bdb5adb4575e1834f811db7fb17ac75344655c3ef3bdbc448e89af16dd2aa66 +size 966844945 diff --git a/pytorch_model-00013-of-00014.bin b/pytorch_model-00013-of-00014.bin new file mode 100644 index 0000000..710bd79 --- /dev/null +++ b/pytorch_model-00013-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac682d1962f4cc89b0ad57471ea140c0d098779805967584b8d40614aded26b8 +size 966828351 diff --git a/pytorch_model-00014-of-00014.bin b/pytorch_model-00014-of-00014.bin new file mode 100644 index 0000000..635e981 --- /dev/null +++ b/pytorch_model-00014-of-00014.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:27e38603cdbdc7612e822396e10ae71c5b1bc458a21ea8a102b3d6689e99a7ad +size 847304239 diff --git a/pytorch_model.bin.index.json b/pytorch_model.bin.index.json new file mode 100644 index 0000000..5d9b525 --- /dev/null +++ b/pytorch_model.bin.index.json @@ -0,0 +1,330 @@ +{ + "metadata": { + "total_size": 13476851712 + }, + "weight_map": { + "lm_head.weight": "pytorch_model-00014-of-00014.bin", + "model.embed_tokens.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.0.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00014.bin", + "model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.1.input_layernorm.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.1.mlp.up_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.1.post_attention_layernorm.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.1.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00014.bin", + "model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00014.bin", + "model.layers.10.input_layernorm.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.10.mlp.down_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.10.mlp.up_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.10.post_attention_layernorm.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.10.self_attn.k_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.10.self_attn.q_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.10.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00014.bin", + "model.layers.10.self_attn.v_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.11.input_layernorm.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.11.mlp.down_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.11.mlp.up_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.11.post_attention_layernorm.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.11.self_attn.k_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.11.self_attn.q_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.11.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00014.bin", + "model.layers.11.self_attn.v_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.12.input_layernorm.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.12.mlp.down_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.12.mlp.up_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.12.post_attention_layernorm.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.12.self_attn.k_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.12.self_attn.q_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.12.self_attn.rotary_emb.inv_freq": "pytorch_model-00006-of-00014.bin", + "model.layers.12.self_attn.v_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.13.input_layernorm.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.13.mlp.down_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.13.mlp.up_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.13.post_attention_layernorm.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.13.self_attn.k_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.13.self_attn.q_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.13.self_attn.rotary_emb.inv_freq": "pytorch_model-00006-of-00014.bin", + "model.layers.13.self_attn.v_proj.weight": "pytorch_model-00006-of-00014.bin", + "model.layers.14.input_layernorm.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.14.mlp.down_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.14.mlp.up_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.14.post_attention_layernorm.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.14.self_attn.k_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.14.self_attn.q_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.14.self_attn.rotary_emb.inv_freq": "pytorch_model-00007-of-00014.bin", + "model.layers.14.self_attn.v_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.15.input_layernorm.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.15.mlp.down_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.15.mlp.up_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.15.post_attention_layernorm.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.15.self_attn.k_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.15.self_attn.q_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.15.self_attn.rotary_emb.inv_freq": "pytorch_model-00007-of-00014.bin", + "model.layers.15.self_attn.v_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.16.input_layernorm.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.16.mlp.down_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.16.mlp.up_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.16.post_attention_layernorm.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.16.self_attn.k_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.16.self_attn.q_proj.weight": "pytorch_model-00007-of-00014.bin", + "model.layers.16.self_attn.rotary_emb.inv_freq": "pytorch_model-00008-of-00014.bin", + "model.layers.16.self_attn.v_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.17.input_layernorm.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.17.mlp.down_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.17.mlp.up_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.17.post_attention_layernorm.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.17.self_attn.k_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.17.self_attn.q_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.17.self_attn.rotary_emb.inv_freq": "pytorch_model-00008-of-00014.bin", + "model.layers.17.self_attn.v_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.18.input_layernorm.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.18.mlp.down_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.18.mlp.up_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.18.post_attention_layernorm.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.18.self_attn.k_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.18.self_attn.q_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.18.self_attn.rotary_emb.inv_freq": "pytorch_model-00008-of-00014.bin", + "model.layers.18.self_attn.v_proj.weight": "pytorch_model-00008-of-00014.bin", + "model.layers.19.input_layernorm.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.19.mlp.down_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.19.mlp.up_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.19.post_attention_layernorm.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.19.self_attn.k_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.19.self_attn.q_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.19.self_attn.rotary_emb.inv_freq": "pytorch_model-00009-of-00014.bin", + "model.layers.19.self_attn.v_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.2.input_layernorm.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.2.mlp.down_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.2.mlp.up_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.2.post_attention_layernorm.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.2.self_attn.k_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.2.self_attn.q_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.2.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00014.bin", + "model.layers.2.self_attn.v_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.20.input_layernorm.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.20.mlp.down_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.20.mlp.up_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.20.post_attention_layernorm.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.20.self_attn.k_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.20.self_attn.q_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.20.self_attn.rotary_emb.inv_freq": "pytorch_model-00009-of-00014.bin", + "model.layers.20.self_attn.v_proj.weight": "pytorch_model-00009-of-00014.bin", + "model.layers.21.input_layernorm.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.21.mlp.down_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.21.mlp.up_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.21.post_attention_layernorm.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.21.self_attn.k_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.21.self_attn.q_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.21.self_attn.rotary_emb.inv_freq": "pytorch_model-00010-of-00014.bin", + "model.layers.21.self_attn.v_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.22.input_layernorm.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.22.mlp.down_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.22.mlp.up_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.22.post_attention_layernorm.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.22.self_attn.k_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.22.self_attn.q_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.22.self_attn.rotary_emb.inv_freq": "pytorch_model-00010-of-00014.bin", + "model.layers.22.self_attn.v_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.23.input_layernorm.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.23.mlp.down_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.23.mlp.up_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.23.post_attention_layernorm.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.23.self_attn.k_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.23.self_attn.q_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.23.self_attn.rotary_emb.inv_freq": "pytorch_model-00010-of-00014.bin", + "model.layers.23.self_attn.v_proj.weight": "pytorch_model-00010-of-00014.bin", + "model.layers.24.input_layernorm.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.24.mlp.down_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.24.mlp.up_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.24.post_attention_layernorm.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.24.self_attn.k_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.24.self_attn.q_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.24.self_attn.rotary_emb.inv_freq": "pytorch_model-00011-of-00014.bin", + "model.layers.24.self_attn.v_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.25.input_layernorm.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.25.mlp.down_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.25.mlp.up_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.25.post_attention_layernorm.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.25.self_attn.k_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.25.self_attn.q_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.25.self_attn.rotary_emb.inv_freq": "pytorch_model-00011-of-00014.bin", + "model.layers.25.self_attn.v_proj.weight": "pytorch_model-00011-of-00014.bin", + "model.layers.26.input_layernorm.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.26.mlp.down_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.26.mlp.up_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.26.post_attention_layernorm.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.26.self_attn.k_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.26.self_attn.q_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.26.self_attn.rotary_emb.inv_freq": "pytorch_model-00012-of-00014.bin", + "model.layers.26.self_attn.v_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.27.input_layernorm.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.27.mlp.down_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.27.mlp.up_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.27.post_attention_layernorm.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.27.self_attn.k_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.27.self_attn.q_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.27.self_attn.rotary_emb.inv_freq": "pytorch_model-00012-of-00014.bin", + "model.layers.27.self_attn.v_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.28.input_layernorm.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.28.mlp.down_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.28.mlp.up_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.28.post_attention_layernorm.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.28.self_attn.k_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.28.self_attn.q_proj.weight": "pytorch_model-00012-of-00014.bin", + "model.layers.28.self_attn.rotary_emb.inv_freq": "pytorch_model-00013-of-00014.bin", + "model.layers.28.self_attn.v_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.29.input_layernorm.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.29.mlp.down_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.29.mlp.up_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.29.post_attention_layernorm.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.29.self_attn.k_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.29.self_attn.q_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.29.self_attn.rotary_emb.inv_freq": "pytorch_model-00013-of-00014.bin", + "model.layers.29.self_attn.v_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.3.input_layernorm.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.3.mlp.down_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.3.mlp.up_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.3.post_attention_layernorm.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.3.self_attn.k_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.3.self_attn.q_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.3.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00014.bin", + "model.layers.3.self_attn.v_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.30.input_layernorm.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.30.mlp.down_proj.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.30.mlp.up_proj.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.30.post_attention_layernorm.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.30.self_attn.k_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.30.self_attn.q_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.30.self_attn.rotary_emb.inv_freq": "pytorch_model-00013-of-00014.bin", + "model.layers.30.self_attn.v_proj.weight": "pytorch_model-00013-of-00014.bin", + "model.layers.31.input_layernorm.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.31.mlp.down_proj.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.31.mlp.up_proj.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.31.post_attention_layernorm.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.31.self_attn.k_proj.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.31.self_attn.q_proj.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.31.self_attn.rotary_emb.inv_freq": "pytorch_model-00014-of-00014.bin", + "model.layers.31.self_attn.v_proj.weight": "pytorch_model-00014-of-00014.bin", + "model.layers.4.input_layernorm.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.4.mlp.down_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.4.mlp.up_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.4.post_attention_layernorm.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.4.self_attn.k_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.4.self_attn.q_proj.weight": "pytorch_model-00002-of-00014.bin", + "model.layers.4.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00014.bin", + "model.layers.4.self_attn.v_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.5.input_layernorm.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.5.mlp.down_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.5.mlp.up_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.5.post_attention_layernorm.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.5.self_attn.k_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.5.self_attn.q_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.5.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00014.bin", + "model.layers.5.self_attn.v_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.6.input_layernorm.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.6.mlp.down_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.6.mlp.up_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.6.post_attention_layernorm.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.6.self_attn.k_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.6.self_attn.q_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.6.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00014.bin", + "model.layers.6.self_attn.v_proj.weight": "pytorch_model-00003-of-00014.bin", + "model.layers.7.input_layernorm.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.7.mlp.down_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.7.mlp.up_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.7.post_attention_layernorm.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.7.self_attn.k_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.7.self_attn.q_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.7.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00014.bin", + "model.layers.7.self_attn.v_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.8.input_layernorm.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.8.mlp.down_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.8.mlp.up_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.8.post_attention_layernorm.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.8.self_attn.k_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.8.self_attn.q_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.8.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00014.bin", + "model.layers.8.self_attn.v_proj.weight": "pytorch_model-00004-of-00014.bin", + "model.layers.9.input_layernorm.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.9.mlp.down_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.9.mlp.up_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.9.post_attention_layernorm.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.9.self_attn.k_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.9.self_attn.q_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.layers.9.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00014.bin", + "model.layers.9.self_attn.v_proj.weight": "pytorch_model-00005-of-00014.bin", + "model.norm.weight": "pytorch_model-00014-of-00014.bin" + } +}