Upload model weights (#1)

- 4c913e5de23d080769591b557e3eee69149b0b4acde1d1e1e87f543df2e0dcdd (735b49878183aab575d8d61ea66b60ee2815421e)
- 987a1e131f533f1a29958279aff540807b0b98adaed05cff99f97ba5193e5d4b (e44be03f16f542cea3b3567659b8816f0887121c)
- 7fd55e0ec21f526d4bb345963ee0b12e779efaf0869996b95f475907a1dfb28c (1f405c7d563bb6256df832541f9bb438f8517e91)
- a561c7e8907bbf12bddc1398f9207c4a72fd1343042a4e92628c06cf912012f5 (bb811686dff6375ff35f4333fd3b78d98d9bb00d)
- 5943cbf7b26103ea8685324a51f42290b06795a77ff5ead5d55980f4e3bd4c0d (71fb28e031c688f8f20b4084bb1f4773b970c0fe)
- 8e351173fc005b26e8eb53486a6dad88e6cbef0f17d6648a97b4f607d89f7bc9 (f4bc14b35c9fba348859058d77e410dd086c4f7e)
- 4882eaa6dc47acb3054f39d3d139e0c4fc8dce394f5a94985a35bedb13e9cf67 (9e5ecd3461995c50b8d9c4a79711de29014d64f7)
- 127c86a8a6faa4f91fafe9aebcf320acfc6901935d172e6a8df7f2311857b79f (1d2c73788e1642a2954ccdc15f33234a0106233d)
- 009f9c6f6c937e431de07018d31a271815405d25f250d322096fac15fa0a4566 (c3f51a7e4ad20307b498ecf42d31772a62564dd8)
- d1f026dca8096f29255ff71e5d117dd6222b5f4e8beb5ecef89fdc56b0cd99a4 (285dae804bfb892970a3b07a7539a8e6a6ebf79c)
- 6aee1ff308bd827022d7c18add85ba2bece325be5401006f495d8292ddddbeae (98fde704ae616afbd0a14f984182694e37473822)
- 8b5e7c94225d0244f4fa56576833aa5836e696338d4a869d3e8a4aff825b2057 (51daf31a55c3dd86c9f0a5f5510fff784a01d509)
- 7c3e4fc9f72bece15de970be690597006dda076afa317666f9a5721563c7a7fb (69e4d805a6bb4dfd0e12ec881f73997e161fbdf5)
- e3d85a94cda627fb7e305686202dcee72ab344947be005f2e7ef684c6d19eadd (4b6345e65c2b2e472121df7f75a590d2680e0f5e)
- cc8efc173d75a81f7c59337241f65720509dc618cf276193c0f12e44c24e84c9 (f3a3c28af222128653917fbc0524a611831934f3)
- 02014f48678bdfb8866a185dc03511c765807a609a038567201bb8f19b15c65d (b76c1d8e73e76dc03339d4a59e51a4ae5a8516fc)
- 5d8d2a7f439619a98f3b88b2497872268a5b2e91ca60c6426a563ed98cff15ee (4f4e6caae7aa4624266072cb41298a631ef7664f)
- 62541edb1b33cfe958da0bdcc68aab4ef8f79d2f484289dfd3fd4d0e1fe356da (e8c3a24585c7e61c2dc374f9c4f7e0b2e8ded54e)
- b5ed337724e5a81d99a79a164715445f6854e4d1660b6a8ca5d8306d98863575 (0dbd5f6f1523e55494c1002ddaac6624e9170047)
- 5f985f8e1a06fa416f98b32214677babb17b2d26f1809a0906a47d5b309c16d4 (a777f5f9b5afd6686a767e493f747411b4823385)
- 2f00d4f1a0e23807a16e53c7c2ad0c410c6a121e72b9f802f4d089c5e8227af4 (d17e7560aca892e8b5b1f62af50f429b3c8d1a92)
- aeec460c798062f4d8f6f0a5c1df58f8c744f51d75708ce8eb7549f875c65560 (9d6ef82700f4fa98138e884693abda8ebcf94b03)

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "/home/llm/mergequant/models/numinous-70b-v1.1-lora",
+  "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": 120,
+  "num_key_value_heads": 8,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.35.2",
+  "use_cache": true,
+  "vocab_size": 32000
+}

model-00001-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2857d3b8818e36ff9a18f0d31952c94d3b21da0cb16234b24bdf51545a620dd4
+size 9550615952

model-00002-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:becded466f66fd0b344a8223604f02c52a1ec603f34fa32c54bb6e5089493128
+size 9949075552

model-00003-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b3cb555b85c3f5891f01308f5fa4504143375f81211d00d684215748a6b70c0
+size 9965868720

model-00004-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f49a7d239b24155aadfd80d9dd2514ba9dc70be87bd81f970157a94d7cd3736b
+size 9647101728

model-00005-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88d1485a8459c437f33399aeb9913f3dfc37177a7dc8502318b3e00749b86727
+size 9798096912

model-00006-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4154e36816864b750a28274db546aa28e05073f33b2b74e4f43ecdc16b2dc1d7
+size 9798096912

model-00007-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3de88bcca7c4a60e28df2a05979cfc28751ba2300574b52d548cd90e660f7ffb
+size 9965835736

model-00008-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73c26e396f386e5a4143c3fcb838c8c0107dc54c8f481a1294300055aa544369
+size 9999423392

model-00009-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13116566d2f8d86f7b9ff3fdca7c2fc9b1ec208a93e027b647146c3c70d77926
+size 9764509240

model-00010-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be1221d8430b8bfb2c3347950d57cf97e1a2bdc24a810ee17a981b14d4e74cdf
+size 9798096904

model-00011-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4b8e79def665ec3a3bfd1808ce42f2abd8546085dd79c76108e24b9ece1891c
+size 9798096912

model-00012-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e4dbf4a0c67e298aae4b11694a6ef43e2cca52dadf00fb5b45eac0806a087c
+size 9798096912

model-00013-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8c12bef67729854efef05c4077ba63304cd8607c77d54981f8ef6563aa44f34
+size 9932315464

model-00014-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57f2396c647e70e137c045fe8b9f6834f932a06d720ad5ebf0fc1ccae100438a
+size 9663878368

model-00015-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6a2939950c028b92919c43d2e0337906d645d8245872923b165a4f0b3a45047
+size 9965835736

model-00016-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d9b8c7d0ba49ece22d48b7d9226212f101cb417b0c9f0a269ebee144a5c0bb3
+size 9999423392

model-00017-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62a6ed719175aa97b5f237eb267a68219f5c1c91d230184e2111eddf29178c1c
+size 9764509240

model-00018-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db54ccdff063fd62a4d9adb42e5fe44ae8b34e285c3aa19f31fc41e16c1c1bd0
+size 9630341632

model-00019-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f05a1b2e2f2191c11d7e490707f915dc8ce56ef593229afce8052a4d95971236
+size 9982629616

model-00020-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da1526a9b32072dd896748b2b8c59106c7f4231f8058a094187a75cd6f39042e
+size 9781303128

model-00021-of-00021.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:747c0567b1699cc6e43a1fd2d675117702bd90f0836e8f7a589478e0913fb243
+size 9852622880

model.safetensors.index.json

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

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "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 +1,39 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<unk>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}