Upload 7 files

config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "Sao10K/Euryale-1.3-L2-70B",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 28672,
+  "max_position_embeddings": 4096,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 140,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.36.0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

huggingface-metadata.txt

@@ -0,0 +1,30 @@
+url: https://huggingface.co/nsfwthrowitaway69/Venus-120b-v1.1
+branch: main
+download date: 2023-12-14 12:07:22
+sha256sum:
+    b8a4a0d2bd814b2c3c533530ee625dab19ff348ae1ada79a673c3787e2fb980d model-00001-of-00025.safetensors
+    ac0c914fc0468cd548f9d003884a47f2b0e75cd16094b8d4eb387f17079e1442 model-00002-of-00025.safetensors
+    47f57bd4111b19ec1b09b10d1d76443d9c3f73f70b8d292d9ad09a2dc1c31f28 model-00003-of-00025.safetensors
+    045015d287ef359d086d4fd70f78ba718ca31e25e6cfa36f54424866228e6270 model-00004-of-00025.safetensors
+    5d256816c5b35fff712d88effcb7cc53baeae1d0601b6228ce929151f4f0f43d model-00005-of-00025.safetensors
+    d2abfec3874134e3f5499269d5bd9306cb73840f759126e6211b2d2de113d416 model-00006-of-00025.safetensors
+    363d71a9680a3e8576e210fb12a736410f715b9e72592fc4b17c9930c410dffb model-00007-of-00025.safetensors
+    e1c51ad1d7911e2cbbac8eb11535a699b472dd4221e01a8e59e6204a2137af62 model-00008-of-00025.safetensors
+    b39f50a9c77eb734a8662166d4db13bce8523f0d83f663446093f4ea675ae8c8 model-00009-of-00025.safetensors
+    f716efc4fe319a2d1885eec56cc2edc52fb01f114cc3cc7982ebbe56a51277d6 model-00010-of-00025.safetensors
+    e0b4d47d54028f6e1c1d8663eef10fa19abfd5c45babf0cdc8b124c16e10a750 model-00011-of-00025.safetensors
+    adc04ad58f86bda3d61ee3129d8811549c26255f0f6b9e16d2d0193fc73d4f46 model-00012-of-00025.safetensors
+    c682b4554f1622061a5c5be9a674d0c7214e8d2a03221a30e046a5f955b91d57 model-00013-of-00025.safetensors
+    56caa5d9d4900a94503c2a0c4029e27692335a4447101c929d56a96ebad58bea model-00014-of-00025.safetensors
+    77dc90321e50cbdc3d0d5a881702faaca59d01493a13d4de29bc4abcbe0abdb5 model-00015-of-00025.safetensors
+    b3a1b7a0798de85305eb9f67580bfeb9357efe9d70e22f4d1e075bb1c8c0ca29 model-00016-of-00025.safetensors
+    553af5c35086dc334423ac709e29a206a6416493aac29f4c3ccc19310f66484e model-00017-of-00025.safetensors
+    67e84f961afe66c144dfd851045bb65c772f15ca2b58e18151c917598157c063 model-00018-of-00025.safetensors
+    9df2aeb203c236f41c64f154c1310fa347194f997acb5d16451109e45e2cc244 model-00019-of-00025.safetensors
+    d1ac06b80cb9ae6e47c333e099d585c5d2d399f409b464b57fbe838a604c3759 model-00020-of-00025.safetensors
+    85d09d823ec8edbe3eb6fff94d5904605bad37b294dd1bc907a0481cc27408ad model-00021-of-00025.safetensors
+    554cf35669b3b0eea9d070b6c1c5b488cf98a557a8f5d296c56e12e3b92295a7 model-00022-of-00025.safetensors
+    2763feb5c1ebb42177449a0b5265d59d2a7286388b9303b2a45726f8efade94c model-00023-of-00025.safetensors
+    a4da530c9e00d17d8e0daf8dfa0e413ccd5f42dc0ab532aed995cfa9770c5f4f model-00024-of-00025.safetensors
+    260a53d93b598f59180c6c8a8a61a84fd224f6b6d9d3f09c3457e86957f77da7 model-00025-of-00025.safetensors
+    9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347 tokenizer.model

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.3.1"}, "weight_map": {"model.embed_tokens.weight": "model-00001-of-00025.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.10.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.20.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.5.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.6.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.8.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.9.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.11.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.21.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.12.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.22.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.13.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.23.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.24.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.25.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.16.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.26.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.17.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.27.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.18.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.28.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.19.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.29.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.30.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.31.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.32.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.33.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.34.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.35.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.36.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.39.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.60.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.61.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.62.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.63.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.64.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.65.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.66.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.67.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.68.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.69.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.70.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.80.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.80.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.80.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.80.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.80.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.80.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.80.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.80.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.80.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.81.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.81.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.81.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.81.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.81.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.81.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.81.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.81.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.81.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.82.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.82.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.82.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.82.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.82.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.82.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.82.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.82.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.82.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.73.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.83.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.83.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.83.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.83.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.83.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.83.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.83.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.83.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.83.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.74.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.84.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.84.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.74.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.84.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.84.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.84.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.84.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.84.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.84.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.84.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.85.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.85.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.85.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.85.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.85.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.85.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.40.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.41.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.75.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.85.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.85.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.85.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.76.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.86.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.86.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.86.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.86.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.86.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.86.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.86.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.86.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.86.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.77.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.87.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.87.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.87.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.87.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.87.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.87.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.87.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.87.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.87.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.78.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.88.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.88.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.88.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.88.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.88.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.88.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.88.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.88.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.88.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.79.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.89.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.89.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.89.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.89.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.89.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.89.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.89.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.89.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.89.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.90.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.100.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.90.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.100.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.90.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.100.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.90.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.100.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.90.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.100.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.90.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.100.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.90.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.100.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.90.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.100.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.90.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.100.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.91.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.101.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.91.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.101.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.91.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.101.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.91.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.101.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.91.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.101.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.42.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.43.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.44.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.91.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.101.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.91.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.101.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.91.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.101.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.91.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.101.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.92.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.102.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.92.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.102.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.92.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.102.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.92.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.102.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.92.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.102.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.92.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.102.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.92.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.102.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.92.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.102.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.92.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.102.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.93.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.103.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.93.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.103.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.93.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.103.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.93.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.103.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.93.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.103.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.93.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.103.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.93.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.103.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.93.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.103.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.93.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.103.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.104.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.94.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.104.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.104.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.104.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.104.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.94.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.104.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.104.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.104.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.104.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.95.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.105.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.95.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.105.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.95.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.105.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.95.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.105.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.95.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.105.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.95.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.105.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.95.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.105.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.95.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.105.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.95.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.105.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.106.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.96.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.106.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.106.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.106.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.106.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.96.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.106.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.106.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.106.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.106.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.97.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.107.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.97.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.107.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.97.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.107.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.97.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.107.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.45.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.46.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.97.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.107.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.97.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.107.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.97.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.107.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.97.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.107.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.97.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.107.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.98.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.108.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.98.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.108.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.108.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.108.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.108.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.98.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.108.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.108.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.108.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.108.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.109.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.99.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.109.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.109.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.109.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.109.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.99.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.109.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.109.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.99.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.109.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.99.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.109.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.110.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.110.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.110.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.110.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.110.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.110.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.110.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.110.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.110.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.111.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.111.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.111.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.111.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.111.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.111.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.111.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.111.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.111.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.112.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.112.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.112.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.112.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.112.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.112.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.112.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.112.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.112.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.47.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.48.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.49.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.113.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.113.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.113.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.113.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.113.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.113.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.113.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.113.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.113.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.114.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.114.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.114.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.114.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.114.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.114.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.114.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.114.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.114.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.115.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.115.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.115.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.115.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.115.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.115.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.115.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.115.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.115.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.116.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.116.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.116.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.116.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.116.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.116.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.116.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.116.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.116.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.117.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.117.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.117.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.117.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.117.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.117.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.117.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.117.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.117.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.118.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.118.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.118.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.118.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.118.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.118.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.50.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.51.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.52.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.118.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.118.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.118.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.119.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.119.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.119.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.119.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.119.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.119.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.119.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.119.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.119.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.53.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.54.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.55.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.56.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.57.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.58.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.59.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.120.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.120.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.120.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.120.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.120.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.121.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.122.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.122.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.122.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.123.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.123.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.124.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.124.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.124.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.124.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.124.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.124.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.124.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.124.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.124.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.125.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.125.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.126.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.126.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.126.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.126.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.126.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.126.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.126.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.126.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.126.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.127.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.127.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.128.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.128.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.129.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.130.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.130.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.130.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.131.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.131.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.132.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.132.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.132.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.132.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.132.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.132.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.132.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.132.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.132.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.133.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.133.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.133.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.134.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.134.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.135.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.136.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.136.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.136.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.136.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.136.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.136.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.136.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.136.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.136.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.137.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.137.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.self_attn.o_proj.weight": "model-00025-of-00025.safetensors", "model.layers.137.self_attn.q_proj.weight": "model-00025-of-00025.safetensors", "model.layers.137.self_attn.v_proj.weight": "model-00025-of-00025.safetensors", "model.layers.138.self_attn.k_proj.weight": "model-00025-of-00025.safetensors", "model.layers.138.self_attn.q_proj.weight": "model-00025-of-00025.safetensors", "model.layers.138.self_attn.v_proj.weight": "model-00025-of-00025.safetensors", "lm_head.weight": "model-00025-of-00025.safetensors", "model.layers.138.input_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.138.mlp.down_proj.weight": "model-00025-of-00025.safetensors", "model.layers.138.mlp.gate_proj.weight": "model-00025-of-00025.safetensors", "model.layers.138.mlp.up_proj.weight": "model-00025-of-00025.safetensors", "model.layers.138.post_attention_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.138.self_attn.o_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.input_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.139.mlp.down_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.mlp.gate_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.mlp.up_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.post_attention_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.139.self_attn.k_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.self_attn.o_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.self_attn.q_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.self_attn.v_proj.weight": "model-00025-of-00025.safetensors", "model.norm.weight": "model-00025-of-00025.safetensors"}}

special_tokens_map.json

@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>"
+  ],
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,45 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>"
+  ],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true
+}