Upload folder using huggingface_hub

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,80 @@
+---
+base_model:
+- Steelskull/L3-MS-Astoria-70b
+- Sao10K/L3-70B-Euryale-v2.1
+library_name: transformers
+tags:
+- mergekit
+- merge
+
+---
+
+# Eurstoria-120B
+
+![](Unholy.png)
+
+A Llama-3 Decoder only model by combining 2x finetuned Llama-3 70B models into 1.
+
+## Merge Details
+### Merge Method
+
+This model was merged using the passthrough merge method.
+
+### Models Merged
+
+The following models were included in the merge:
+* [Steelskull/L3-MS-Astoria-70b](https://huggingface.co/Steelskull/L3-MS-Astoria-70b)
+* [Sao10K/L3-70B-Euryale-v2.1](https://huggingface.co/Sao10K/L3-70B-Euryale-v2.1)
+
+### Configuration
+
+The following YAML configuration was used to produce this model:
+
+```yaml
+slices:
+  - sources:
+    - model: Steelskull/L3-MS-Astoria-70b
+      layer_range: [0, 16]
+  - sources:
+    - model: Sao10K/L3-70B-Euryale-v2.1
+      layer_range: [8, 24]
+  - sources:
+    - model: Steelskull/L3-MS-Astoria-70b
+      layer_range: [25, 40]
+  - sources:
+    - model: Sao10K/L3-70B-Euryale-v2.1
+      layer_range: [33, 48]
+  - sources:
+    - model: Steelskull/L3-MS-Astoria-70b
+      layer_range: [41, 56]
+  - sources:
+    - model: Sao10K/L3-70B-Euryale-v2.1
+      layer_range: [49, 64]
+  - sources:
+    - model: Steelskull/L3-MS-Astoria-70b
+      layer_range: [57, 72]
+  - sources:
+    - model: Sao10K/L3-70B-Euryale-v2.1
+      layer_range: [65, 80]
+
+merge_method: passthrough
+dtype: float16
+```
+
+# Prompting Format
+
+As both models are based off from instruct, Use the default [Llama 3 Instruct format](https://huggingface.co/Sao10K/L3-70B-Euryale-v2.1/blob/main/Euryale-v2.1-Llama-3-Instruct.json) would be fine.
+
+# Screenshots
+
+None. Unfortunately or Fortunately.
+
+# Benchmarks
+
+No one is willing to bench a 120B... right?
+
+# Acknowledgements
+
+The original ratios come from Goliath and are used as is. As such thanks to both [Alpindale](https://huggingface.co/alpindale) and [@Undi95](https://huggingface.co/Undi95) for their inital work.
+
+Credits goes to [@chargoddard](https://huggingface.co/chargoddard) for developing the framework used to merge the model - [mergekit](https://github.com/cg123/mergekit).

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "Steelskull/L3-MS-Astoria-70b",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": 128001,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 28672,
+  "max_position_embeddings": 8192,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 122,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.41.2",
+  "use_cache": true,
+  "vocab_size": 128256
+}

mergekit_config.yml

@@ -0,0 +1,28 @@
+slices:
+  - sources:
+    - model: Steelskull/L3-MS-Astoria-70b
+      layer_range: [0, 16]
+  - sources:
+    - model: Sao10K/L3-70B-Euryale-v2.1
+      layer_range: [8, 24]
+  - sources:
+    - model: Steelskull/L3-MS-Astoria-70b
+      layer_range: [25, 40]
+  - sources:
+    - model: Sao10K/L3-70B-Euryale-v2.1
+      layer_range: [33, 48]
+  - sources:
+    - model: Steelskull/L3-MS-Astoria-70b
+      layer_range: [41, 56]
+  - sources:
+    - model: Sao10K/L3-70B-Euryale-v2.1
+      layer_range: [49, 64]
+  - sources:
+    - model: Steelskull/L3-MS-Astoria-70b
+      layer_range: [57, 72]
+  - sources:
+    - model: Sao10K/L3-70B-Euryale-v2.1
+      layer_range: [65, 80]
+
+merge_method: passthrough
+dtype: float16

model-00001-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41ca4491ea1a6d0a61860b9225cab29dbf9fac063ee92321cc4e185f6e6175da
+size 4672471496

model-00002-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f94625dd202e5c7e76f619464edbe9f70514803040bf1a8542b56818e914510
+size 4831922680

model-00003-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c80b19b64fd0bc758e1d6385305b3d7478e44b19e2c5c7029e30114b4a734ca
+size 4966190736

model-00004-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1a97601abee0eda286e8dcb77ba776ec7cf7d0cc747228e55692826fc3c6209
+size 4831906200

model-00005-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6542f0dfe47a27008d834e53d96b074edcff6eb54e33a52b1a0af67800d29651
+size 4999711712

model-00006-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15680cf1c9dbf72539f666869c1d1373711388026e14c3a9ba0ed7d90138b049
+size 4798368728

model-00007-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd49ba662a42ab9e11d7c74a39fca222155e44df71799a4e0f6b3680c9ddac04
+size 4664167384

model-00008-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9bde4c6cfd2930339c3ab6c5cbd4f569c08760c767fd48495979a333babf57fd
+size 4664167376

model-00009-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a5d1007bfc3f621d9a905238ca0001217ce8a150d5b35504c4f86b8506ad2e0
+size 4982917880

model-00010-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b13b181c54567bd49cc29a656a0d1ad4b8ba198067110b1b503be5d37eb23262
+size 4999711704

model-00011-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e819f3ebd0156237a61ea5652dc67fbf4b8df53edf35d72646f9a36641964a3a
+size 4647373544

model-00012-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd4147d4002cbe52fba79b7d793a5e705055864b673fff33c08721f86ba4515c
+size 4664167376

model-00013-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40295220572dafdba97e2e85307c404233e47e663a6779b58f43b5007f27992a
+size 4664183888

model-00014-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f9aabf35b9a551353abdd0fe647f0686a55c8cb176fe83eb47a06cfd4d9b3d3
+size 4831906200

model-00015-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3de99752c536ec8d23655caf0abd58391a192fbf9d580fb46346d51a31e1d11e
+size 4999711712

model-00016-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee90ed42b48f029e9a32c0a798b7233066e593d32a8a16e9d8470a321eeee44a
+size 4798385224

model-00017-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:101580b04edd021b6f174cc43f50495e0ba78a9ee1aaad377e761fb3806df827
+size 4831906200

model-00018-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de6f6f926c9a6dd4340636f153d0f8869d64cb8251e423015d7a20a7ec51ca3b
+size 4966190728

model-00019-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ec710fff31dc4a777f99a7c6d60fc10d780625fd9ab0edb108a77f0a89f62b3
+size 4831906200

model-00020-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d21dfd78ee39cde9c086b73f594cc179fe1e58e4e1a49c19c88a30bd5617c40
+size 4966190736

model-00021-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abfa7f6ea4ffe0c53d7d3613f93251a23a991a79af48598fb719f0ef449f6d11
+size 4831906200

model-00022-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b02bd5b022edc2eb6a5a667008188701607f8785a2109927e2a6bdbf7cdd0773
+size 4999711712

model-00023-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ffa5afaffc5c6f9dedcd7d7efd206f8e945011f8e43f30e749e01491cef30e2f
+size 4798368728

model-00024-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8114fe183fb28aaede9087d74accabb2aee1349c50abc3b8a35608750e79f599
+size 4664167376

model-00025-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45ac8d1114734efba34a5bfc945819c0e5e19cc067584eef03d2d541826f6ef3
+size 4664183880

model-00026-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:074441e881073155ddcdc0cac4923890c3d833b6ddbe2903e6d3ac3711cc4db7
+size 4831906200

model-00027-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ca24d7d9e51e4f1a00beb43078140b5be87a3e2779f74e7d222e21d6ae8f23c
+size 4999711712

model-00028-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c4f93bc25b5de70d49796502cb33d590c07090031c7fbcbdb1c77d15d88696b
+size 4798385224

model-00029-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85bad9f58f7773c5165f75a9cab03af86f15d83da76eba7538a3afa14212d693
+size 4664167392

model-00030-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2ac4c612615baf624c9bbb16cfac2718c6cb47cda00cb9546b4d0fe3175bc1c
+size 4831906200

model-00031-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e638521a22c807464a4c9835740766228e6520d99aa5b0d6332cc69ac3ae98cb
+size 4966174232

model-00032-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8884c156325cf3aaea6c7731d01c4743939d928cd22571f71937e3b8859ae73d
+size 4966140544

model-00033-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:592592ebfa374f4ef3adfdfe66e69ef9a98b7596cf12b51f03a2e0b33c9c8e20
+size 4999711704

model-00034-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9171a859eea00e999d379984510db4dfa00ddb45f11bffebd31525fc7089ecf4
+size 4999711712

model-00035-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:651ee0fda96ed266665a2dd8821d52038353dd1b55b8f5a8a8f3ec857f2b5a9c
+size 4999711712

model-00036-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e3e50baf3579ca514aad7d9b20b6ecca2f206cb0bd9063ad60a8ee900984e96
+size 4932603096

model-00037-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0d3f30965bfaeae3b144e5508eae4de9d4e9c951689060d34d14fabb1a5614d
+size 4831906216

model-00038-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0b8df72b6e5aba83f0e0549d9f401c4a84c92cfe04353a1a477933dd86b6bac
+size 4966174272

model-00039-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5fa609b7a225d1ed10d61d4b9d6d9ef15210e53d17ca9c70e69ad184937f826
+size 4664183888

model-00040-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f98b53f3f7a084c7053687da3a000a0e68c2ced9a1d09398b5dce016920d8a95
+size 4982901408

model-00041-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c21db40cbef9e7867779a7b5041492f822a7c98ddb894bdeb323e0396d8639a
+size 4999711736

model-00042-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4b395cdb38ac5294f656e1df56f744576dd4f237dc697a8c53132aceb9f270b
+size 4647373576

model-00043-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:667cefdbac30068c8ff26957f6b051971983334592d7192b6f5a3391293d2452
+size 4831922696

model-00044-of-00044.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d295f4d7af3293fe387b6880772b93d9643539a33fe06d1b1c033f9105a79b67
+size 4496412032

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.4.2", "total_size": 212982382592}, "weight_map": {"lm_head.weight": "model-00001-of-00044.safetensors", "model.embed_tokens.weight": "model-00001-of-00044.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00044.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00001-of-00044.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00002-of-00044.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00002-of-00044.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00002-of-00044.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00002-of-00044.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00002-of-00044.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00002-of-00044.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00002-of-00044.safetensors", "model.layers.1.input_layernorm.weight": "model-00002-of-00044.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00002-of-00044.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00044.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00002-of-00044.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00044.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00044.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00002-of-00044.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00044.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00044.safetensors", "model.layers.18.input_layernorm.weight": "model-00002-of-00044.safetensors", "model.layers.10.input_layernorm.weight": "model-00002-of-00044.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00002-of-00044.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00002-of-00044.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00044.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00044.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00003-of-00044.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00003-of-00044.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00044.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00044.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00003-of-00044.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00044.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00003-of-00044.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00044.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00003-of-00044.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00044.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00003-of-00044.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00044.safetensors", "model.layers.19.input_layernorm.weight": "model-00003-of-00044.safetensors", "model.layers.11.input_layernorm.weight": "model-00003-of-00044.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00003-of-00044.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00003-of-00044.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00044.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00044.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00003-of-00044.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00003-of-00044.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00044.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00044.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00044.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00044.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00044.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00044.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00044.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00044.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00044.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00044.safetensors", "model.layers.20.input_layernorm.weight": "model-00003-of-00044.safetensors", "model.layers.12.input_layernorm.weight": "model-00003-of-00044.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00004-of-00044.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00004-of-00044.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00004-of-00044.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00004-of-00044.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00004-of-00044.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00004-of-00044.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00004-of-00044.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00004-of-00044.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00004-of-00044.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00004-of-00044.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00004-of-00044.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00004-of-00044.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00004-of-00044.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00004-of-00044.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00004-of-00044.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00004-of-00044.safetensors", "model.layers.21.input_layernorm.weight": "model-00004-of-00044.safetensors", "model.layers.13.input_layernorm.weight": "model-00004-of-00044.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00004-of-00044.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00004-of-00044.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00004-of-00044.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00005-of-00044.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00005-of-00044.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00005-of-00044.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00044.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00005-of-00044.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00044.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00005-of-00044.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00044.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00005-of-00044.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00044.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00005-of-00044.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00044.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00005-of-00044.safetensors", "model.layers.22.input_layernorm.weight": "model-00005-of-00044.safetensors", "model.layers.14.input_layernorm.weight": "model-00005-of-00044.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00005-of-00044.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00005-of-00044.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00005-of-00044.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00005-of-00044.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00005-of-00044.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00005-of-00044.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00005-of-00044.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00005-of-00044.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00044.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00005-of-00044.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00044.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00006-of-00044.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00006-of-00044.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00006-of-00044.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00006-of-00044.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00006-of-00044.safetensors", "model.layers.23.input_layernorm.weight": "model-00006-of-00044.safetensors", "model.layers.15.input_layernorm.weight": "model-00006-of-00044.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00006-of-00044.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00006-of-00044.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00006-of-00044.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00006-of-00044.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00006-of-00044.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00006-of-00044.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00044.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00006-of-00044.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00006-of-00044.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00006-of-00044.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00006-of-00044.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00006-of-00044.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00006-of-00044.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00006-of-00044.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00006-of-00044.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00006-of-00044.safetensors", "model.layers.24.input_layernorm.weight": "model-00006-of-00044.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00006-of-00044.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00044.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00007-of-00044.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00007-of-00044.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00007-of-00044.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00007-of-00044.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00007-of-00044.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00007-of-00044.safetensors", "model.layers.25.input_layernorm.weight": "model-00007-of-00044.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00007-of-00044.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00007-of-00044.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00007-of-00044.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00007-of-00044.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00007-of-00044.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00007-of-00044.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00007-of-00044.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00007-of-00044.safetensors", "model.layers.26.input_layernorm.weight": "model-00007-of-00044.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00007-of-00044.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00007-of-00044.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00007-of-00044.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00007-of-00044.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00007-of-00044.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00007-of-00044.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00007-of-00044.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00007-of-00044.safetensors", "model.layers.27.input_layernorm.weight": "model-00007-of-00044.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00007-of-00044.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00008-of-00044.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00008-of-00044.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00008-of-00044.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00008-of-00044.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00008-of-00044.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00008-of-00044.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00008-of-00044.safetensors", "model.layers.2.input_layernorm.weight": "model-00008-of-00044.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00008-of-00044.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00008-of-00044.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00008-of-00044.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00008-of-00044.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00008-of-00044.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00008-of-00044.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00008-of-00044.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00008-of-00044.safetensors", "model.layers.28.input_layernorm.weight": "model-00008-of-00044.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00008-of-00044.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00008-of-00044.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00008-of-00044.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00008-of-00044.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00008-of-00044.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00008-of-00044.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00008-of-00044.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00008-of-00044.safetensors", "model.layers.29.input_layernorm.weight": "model-00008-of-00044.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00009-of-00044.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00009-of-00044.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00009-of-00044.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00009-of-00044.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00009-of-00044.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00009-of-00044.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00009-of-00044.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00009-of-00044.safetensors", "model.layers.30.input_layernorm.weight": "model-00009-of-00044.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00009-of-00044.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00009-of-00044.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00009-of-00044.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00009-of-00044.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00009-of-00044.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00009-of-00044.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00009-of-00044.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00009-of-00044.safetensors", "model.layers.31.input_layernorm.weight": "model-00009-of-00044.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00009-of-00044.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00009-of-00044.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00009-of-00044.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00009-of-00044.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00009-of-00044.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00009-of-00044.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00010-of-00044.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00010-of-00044.safetensors", "model.layers.32.input_layernorm.weight": "model-00010-of-00044.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00010-of-00044.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00010-of-00044.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00010-of-00044.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00010-of-00044.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00010-of-00044.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00010-of-00044.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00010-of-00044.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00010-of-00044.safetensors", "model.layers.33.input_layernorm.weight": "model-00010-of-00044.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00010-of-00044.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00010-of-00044.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00010-of-00044.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00010-of-00044.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00010-of-00044.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00010-of-00044.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00010-of-00044.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00010-of-00044.safetensors", "model.layers.34.input_layernorm.weight": "model-00010-of-00044.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00010-of-00044.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00010-of-00044.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00010-of-00044.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00010-of-00044.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00010-of-00044.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00011-of-00044.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00011-of-00044.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00011-of-00044.safetensors", "model.layers.35.input_layernorm.weight": "model-00011-of-00044.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00011-of-00044.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00011-of-00044.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00011-of-00044.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00011-of-00044.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00011-of-00044.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00011-of-00044.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00011-of-00044.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00011-of-00044.safetensors", "model.layers.36.input_layernorm.weight": "model-00011-of-00044.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00011-of-00044.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00011-of-00044.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00011-of-00044.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00011-of-00044.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00011-of-00044.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00011-of-00044.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00011-of-00044.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00011-of-00044.safetensors", "model.layers.3.input_layernorm.weight": "model-00011-of-00044.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00011-of-00044.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00011-of-00044.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00012-of-00044.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00012-of-00044.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00012-of-00044.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00012-of-00044.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00012-of-00044.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00012-of-00044.safetensors", "model.layers.37.input_layernorm.weight": "model-00012-of-00044.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00012-of-00044.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00012-of-00044.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00012-of-00044.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00012-of-00044.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00012-of-00044.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00012-of-00044.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00012-of-00044.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00012-of-00044.safetensors", "model.layers.38.input_layernorm.weight": "model-00012-of-00044.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00012-of-00044.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00012-of-00044.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00012-of-00044.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00012-of-00044.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00012-of-00044.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00012-of-00044.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00012-of-00044.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00012-of-00044.safetensors", "model.layers.39.input_layernorm.weight": "model-00012-of-00044.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00012-of-00044.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00013-of-00044.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00013-of-00044.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00013-of-00044.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00013-of-00044.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00013-of-00044.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00013-of-00044.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00013-of-00044.safetensors", "model.layers.47.input_layernorm.weight": "model-00013-of-00044.safetensors", "model.layers.40.input_layernorm.weight": "model-00013-of-00044.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00013-of-00044.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00013-of-00044.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00013-of-00044.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00013-of-00044.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00013-of-00044.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00013-of-00044.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00013-of-00044.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00013-of-00044.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00013-of-00044.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00013-of-00044.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00013-of-00044.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00013-of-00044.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00013-of-00044.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00013-of-00044.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00013-of-00044.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00013-of-00044.safetensors", "model.layers.48.input_layernorm.weight": "model-00013-of-00044.safetensors", "model.layers.41.input_layernorm.weight": "model-00013-of-00044.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00014-of-00044.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00014-of-00044.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00014-of-00044.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00014-of-00044.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00014-of-00044.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00014-of-00044.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00014-of-00044.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00014-of-00044.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00014-of-00044.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00014-of-00044.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00014-of-00044.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00014-of-00044.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00014-of-00044.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00014-of-00044.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00014-of-00044.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00014-of-00044.safetensors", "model.layers.49.input_layernorm.weight": "model-00014-of-00044.safetensors", "model.layers.42.input_layernorm.weight": "model-00014-of-00044.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00014-of-00044.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00014-of-00044.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00014-of-00044.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00015-of-00044.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00015-of-00044.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00015-of-00044.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00015-of-00044.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00015-of-00044.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00015-of-00044.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00015-of-00044.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00015-of-00044.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00015-of-00044.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00015-of-00044.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00015-of-00044.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00015-of-00044.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00015-of-00044.safetensors", "model.layers.50.input_layernorm.weight": "model-00015-of-00044.safetensors", "model.layers.43.input_layernorm.weight": "model-00015-of-00044.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00015-of-00044.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00015-of-00044.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00015-of-00044.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00015-of-00044.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00015-of-00044.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00015-of-00044.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00015-of-00044.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00015-of-00044.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00015-of-00044.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00015-of-00044.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00015-of-00044.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00016-of-00044.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00016-of-00044.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00016-of-00044.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00016-of-00044.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00016-of-00044.safetensors", "model.layers.51.input_layernorm.weight": "model-00016-of-00044.safetensors", "model.layers.44.input_layernorm.weight": "model-00016-of-00044.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00016-of-00044.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00016-of-00044.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00016-of-00044.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00016-of-00044.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00016-of-00044.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00016-of-00044.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00016-of-00044.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00016-of-00044.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00016-of-00044.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00016-of-00044.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00016-of-00044.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00016-of-00044.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00016-of-00044.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00016-of-00044.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00016-of-00044.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00016-of-00044.safetensors", "model.layers.52.input_layernorm.weight": "model-00016-of-00044.safetensors", "model.layers.45.input_layernorm.weight": "model-00016-of-00044.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00016-of-00044.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00016-of-00044.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00017-of-00044.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00017-of-00044.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00017-of-00044.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00017-of-00044.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00017-of-00044.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00017-of-00044.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00017-of-00044.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00017-of-00044.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00017-of-00044.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00017-of-00044.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00017-of-00044.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00017-of-00044.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00017-of-00044.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00017-of-00044.safetensors", "model.layers.53.input_layernorm.weight": "model-00017-of-00044.safetensors", "model.layers.46.input_layernorm.weight": "model-00017-of-00044.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00017-of-00044.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00017-of-00044.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00017-of-00044.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00017-of-00044.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00017-of-00044.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00018-of-00044.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00018-of-00044.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00018-of-00044.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00018-of-00044.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00018-of-00044.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00018-of-00044.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00018-of-00044.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00018-of-00044.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00018-of-00044.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00018-of-00044.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00018-of-00044.safetensors", "model.layers.4.input_layernorm.weight": "model-00018-of-00044.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00018-of-00044.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00018-of-00044.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00018-of-00044.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00018-of-00044.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00018-of-00044.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00018-of-00044.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00018-of-00044.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00018-of-00044.safetensors", "model.layers.54.input_layernorm.weight": "model-00018-of-00044.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00018-of-00044.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00018-of-00044.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00018-of-00044.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00018-of-00044.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00018-of-00044.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00018-of-00044.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00018-of-00044.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00018-of-00044.safetensors", "model.layers.62.input_layernorm.weight": "model-00018-of-00044.safetensors", "model.layers.55.input_layernorm.weight": "model-00018-of-00044.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00018-of-00044.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00019-of-00044.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00019-of-00044.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00019-of-00044.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00019-of-00044.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00019-of-00044.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00019-of-00044.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00019-of-00044.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00019-of-00044.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00019-of-00044.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00019-of-00044.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00019-of-00044.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00019-of-00044.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00019-of-00044.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00019-of-00044.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00019-of-00044.safetensors", "model.layers.63.input_layernorm.weight": "model-00019-of-00044.safetensors", "model.layers.56.input_layernorm.weight": "model-00019-of-00044.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00019-of-00044.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00019-of-00044.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00019-of-00044.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00019-of-00044.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00020-of-00044.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00020-of-00044.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00020-of-00044.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00020-of-00044.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00020-of-00044.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00020-of-00044.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00020-of-00044.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00020-of-00044.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00020-of-00044.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00020-of-00044.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00020-of-00044.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00020-of-00044.safetensors", "model.layers.64.input_layernorm.weight": "model-00020-of-00044.safetensors", "model.layers.57.input_layernorm.weight": "model-00020-of-00044.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00020-of-00044.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00020-of-00044.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00020-of-00044.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00020-of-00044.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00020-of-00044.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00020-of-00044.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00020-of-00044.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00020-of-00044.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00020-of-00044.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00020-of-00044.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00020-of-00044.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00020-of-00044.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00020-of-00044.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00020-of-00044.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00020-of-00044.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00020-of-00044.safetensors", "model.layers.65.input_layernorm.weight": "model-00020-of-00044.safetensors", "model.layers.58.input_layernorm.weight": "model-00020-of-00044.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00021-of-00044.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00021-of-00044.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00021-of-00044.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00021-of-00044.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00021-of-00044.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00021-of-00044.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00021-of-00044.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00021-of-00044.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00021-of-00044.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00021-of-00044.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00021-of-00044.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00021-of-00044.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00021-of-00044.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00021-of-00044.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00021-of-00044.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00021-of-00044.safetensors", "model.layers.66.input_layernorm.weight": "model-00021-of-00044.safetensors", "model.layers.59.input_layernorm.weight": "model-00021-of-00044.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00021-of-00044.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00021-of-00044.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00021-of-00044.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00022-of-00044.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00022-of-00044.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00022-of-00044.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00022-of-00044.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00022-of-00044.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00022-of-00044.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00022-of-00044.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00022-of-00044.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00022-of-00044.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00022-of-00044.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00022-of-00044.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00022-of-00044.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00022-of-00044.safetensors", "model.layers.67.input_layernorm.weight": "model-00022-of-00044.safetensors", "model.layers.60.input_layernorm.weight": "model-00022-of-00044.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00022-of-00044.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00022-of-00044.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00022-of-00044.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00022-of-00044.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00022-of-00044.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00022-of-00044.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00022-of-00044.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00022-of-00044.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00022-of-00044.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00022-of-00044.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00022-of-00044.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00023-of-00044.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00023-of-00044.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00023-of-00044.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00023-of-00044.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00023-of-00044.safetensors", "model.layers.68.input_layernorm.weight": "model-00023-of-00044.safetensors", "model.layers.61.input_layernorm.weight": "model-00023-of-00044.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00023-of-00044.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00023-of-00044.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00023-of-00044.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00023-of-00044.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00023-of-00044.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00023-of-00044.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00023-of-00044.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00023-of-00044.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00023-of-00044.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00023-of-00044.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00023-of-00044.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00023-of-00044.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00023-of-00044.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00023-of-00044.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00023-of-00044.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00023-of-00044.safetensors", "model.layers.69.input_layernorm.weight": "model-00023-of-00044.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00023-of-00044.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00023-of-00044.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00024-of-00044.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00024-of-00044.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00024-of-00044.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00024-of-00044.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00024-of-00044.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00024-of-00044.safetensors", "model.layers.77.input_layernorm.weight": "model-00024-of-00044.safetensors", "model.layers.70.input_layernorm.weight": "model-00024-of-00044.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00024-of-00044.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00024-of-00044.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00024-of-00044.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00024-of-00044.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00024-of-00044.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00024-of-00044.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00024-of-00044.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00024-of-00044.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00024-of-00044.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00024-of-00044.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00024-of-00044.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00024-of-00044.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00024-of-00044.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00024-of-00044.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00024-of-00044.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00024-of-00044.safetensors", "model.layers.5.input_layernorm.weight": "model-00024-of-00044.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00024-of-00044.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00025-of-00044.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00025-of-00044.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00025-of-00044.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00025-of-00044.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00025-of-00044.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00025-of-00044.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00025-of-00044.safetensors", "model.layers.78.input_layernorm.weight": "model-00025-of-00044.safetensors", "model.layers.71.input_layernorm.weight": "model-00025-of-00044.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00025-of-00044.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00025-of-00044.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00025-of-00044.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00025-of-00044.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00025-of-00044.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00025-of-00044.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00025-of-00044.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00025-of-00044.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00025-of-00044.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00025-of-00044.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00025-of-00044.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00025-of-00044.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00025-of-00044.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00025-of-00044.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00025-of-00044.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00025-of-00044.safetensors", "model.layers.79.input_layernorm.weight": "model-00025-of-00044.safetensors", "model.layers.72.input_layernorm.weight": "model-00025-of-00044.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00026-of-00044.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00026-of-00044.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00026-of-00044.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00026-of-00044.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00026-of-00044.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00026-of-00044.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00026-of-00044.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00026-of-00044.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00026-of-00044.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00026-of-00044.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00026-of-00044.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00026-of-00044.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00026-of-00044.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00026-of-00044.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00026-of-00044.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00026-of-00044.safetensors", "model.layers.80.input_layernorm.weight": "model-00026-of-00044.safetensors", "model.layers.73.input_layernorm.weight": "model-00026-of-00044.safetensors", "model.layers.80.mlp.down_proj.weight": "model-00026-of-00044.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00026-of-00044.safetensors", "model.layers.80.mlp.gate_proj.weight": "model-00026-of-00044.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00027-of-00044.safetensors", "model.layers.80.mlp.up_proj.weight": "model-00027-of-00044.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00027-of-00044.safetensors", "model.layers.80.post_attention_layernorm.weight": "model-00027-of-00044.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00027-of-00044.safetensors", "model.layers.80.self_attn.k_proj.weight": "model-00027-of-00044.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00027-of-00044.safetensors", "model.layers.80.self_attn.o_proj.weight": "model-00027-of-00044.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00027-of-00044.safetensors", "model.layers.80.self_attn.q_proj.weight": "model-00027-of-00044.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00027-of-00044.safetensors", "model.layers.80.self_attn.v_proj.weight": "model-00027-of-00044.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00027-of-00044.safetensors", "model.layers.81.input_layernorm.weight": "model-00027-of-00044.safetensors", "model.layers.74.input_layernorm.weight": "model-00027-of-00044.safetensors", "model.layers.81.mlp.down_proj.weight": "model-00027-of-00044.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00027-of-00044.safetensors", "model.layers.81.mlp.gate_proj.weight": "model-00027-of-00044.safetensors", "model.layers.74.mlp.gate_proj.weight": "model-00027-of-00044.safetensors", "model.layers.81.mlp.up_proj.weight": "model-00027-of-00044.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00027-of-00044.safetensors", "model.layers.81.post_attention_layernorm.weight": "model-00027-of-00044.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00027-of-00044.safetensors", "model.layers.81.self_attn.k_proj.weight": "model-00027-of-00044.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00027-of-00044.safetensors", "model.layers.81.self_attn.o_proj.weight": "model-00027-of-00044.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00028-of-00044.safetensors", "model.layers.81.self_attn.q_proj.weight": "model-00028-of-00044.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00028-of-00044.safetensors", "model.layers.81.self_attn.v_proj.weight": "model-00028-of-00044.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00028-of-00044.safetensors", "model.layers.82.input_layernorm.weight": "model-00028-of-00044.safetensors", "model.layers.75.input_layernorm.weight": "model-00028-of-00044.safetensors", "model.layers.82.mlp.down_proj.weight": "model-00028-of-00044.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00028-of-00044.safetensors", "model.layers.82.mlp.gate_proj.weight": "model-00028-of-00044.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00028-of-00044.safetensors", "model.layers.82.mlp.up_proj.weight": "model-00028-of-00044.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00028-of-00044.safetensors", "model.layers.82.post_attention_layernorm.weight": "model-00028-of-00044.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00028-of-00044.safetensors", "model.layers.82.self_attn.k_proj.weight": "model-00028-of-00044.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00028-of-00044.safetensors", "model.layers.82.self_attn.o_proj.weight": "model-00028-of-00044.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00028-of-00044.safetensors", "model.layers.82.self_attn.q_proj.weight": "model-00028-of-00044.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00028-of-00044.safetensors", "model.layers.82.self_attn.v_proj.weight": "model-00028-of-00044.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00028-of-00044.safetensors", "model.layers.83.input_layernorm.weight": "model-00028-of-00044.safetensors", "model.layers.76.input_layernorm.weight": "model-00028-of-00044.safetensors", "model.layers.83.mlp.down_proj.weight": "model-00028-of-00044.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00028-of-00044.safetensors", "model.layers.83.mlp.gate_proj.weight": "model-00029-of-00044.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00029-of-00044.safetensors", "model.layers.83.mlp.up_proj.weight": "model-00029-of-00044.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00029-of-00044.safetensors", "model.layers.83.post_attention_layernorm.weight": "model-00029-of-00044.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00029-of-00044.safetensors", "model.layers.83.self_attn.k_proj.weight": "model-00029-of-00044.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00029-of-00044.safetensors", "model.layers.83.self_attn.o_proj.weight": "model-00029-of-00044.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00029-of-00044.safetensors", "model.layers.83.self_attn.q_proj.weight": "model-00029-of-00044.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00029-of-00044.safetensors", "model.layers.83.self_attn.v_proj.weight": "model-00029-of-00044.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00029-of-00044.safetensors", "model.layers.84.input_layernorm.weight": "model-00029-of-00044.safetensors", "model.layers.84.mlp.down_proj.weight": "model-00029-of-00044.safetensors", "model.layers.84.mlp.gate_proj.weight": "model-00029-of-00044.safetensors", "model.layers.84.mlp.up_proj.weight": "model-00029-of-00044.safetensors", "model.layers.84.post_attention_layernorm.weight": "model-00029-of-00044.safetensors", "model.layers.84.self_attn.k_proj.weight": "model-00029-of-00044.safetensors", "model.layers.84.self_attn.o_proj.weight": "model-00029-of-00044.safetensors", "model.layers.84.self_attn.q_proj.weight": "model-00029-of-00044.safetensors", "model.layers.84.self_attn.v_proj.weight": "model-00029-of-00044.safetensors", "model.layers.92.input_layernorm.weight": "model-00029-of-00044.safetensors", "model.layers.85.input_layernorm.weight": "model-00029-of-00044.safetensors", "model.layers.92.mlp.down_proj.weight": "model-00029-of-00044.safetensors", "model.layers.85.mlp.down_proj.weight": "model-00030-of-00044.safetensors", "model.layers.92.mlp.gate_proj.weight": "model-00030-of-00044.safetensors", "model.layers.85.mlp.gate_proj.weight": "model-00030-of-00044.safetensors", "model.layers.92.mlp.up_proj.weight": "model-00030-of-00044.safetensors", "model.layers.85.mlp.up_proj.weight": "model-00030-of-00044.safetensors", "model.layers.92.post_attention_layernorm.weight": "model-00030-of-00044.safetensors", "model.layers.85.post_attention_layernorm.weight": "model-00030-of-00044.safetensors", "model.layers.92.self_attn.k_proj.weight": "model-00030-of-00044.safetensors", "model.layers.85.self_attn.k_proj.weight": "model-00030-of-00044.safetensors", "model.layers.92.self_attn.o_proj.weight": "model-00030-of-00044.safetensors", "model.layers.85.self_attn.o_proj.weight": "model-00030-of-00044.safetensors", "model.layers.92.self_attn.q_proj.weight": "model-00030-of-00044.safetensors", "model.layers.85.self_attn.q_proj.weight": "model-00030-of-00044.safetensors", "model.layers.92.self_attn.v_proj.weight": "model-00030-of-00044.safetensors", "model.layers.85.self_attn.v_proj.weight": "model-00030-of-00044.safetensors", "model.layers.93.input_layernorm.weight": "model-00030-of-00044.safetensors", "model.layers.86.input_layernorm.weight": "model-00030-of-00044.safetensors", "model.layers.93.mlp.down_proj.weight": "model-00030-of-00044.safetensors", "model.layers.86.mlp.down_proj.weight": "model-00030-of-00044.safetensors", "model.layers.93.mlp.gate_proj.weight": "model-00030-of-00044.safetensors", "model.layers.86.mlp.gate_proj.weight": "model-00030-of-00044.safetensors", "model.layers.93.mlp.up_proj.weight": "model-00031-of-00044.safetensors", "model.layers.86.mlp.up_proj.weight": "model-00031-of-00044.safetensors", "model.layers.93.post_attention_layernorm.weight": "model-00031-of-00044.safetensors", "model.layers.86.post_attention_layernorm.weight": "model-00031-of-00044.safetensors", "model.layers.93.self_attn.k_proj.weight": "model-00031-of-00044.safetensors", "model.layers.86.self_attn.k_proj.weight": "model-00031-of-00044.safetensors", "model.layers.93.self_attn.o_proj.weight": "model-00031-of-00044.safetensors", "model.layers.86.self_attn.o_proj.weight": "model-00031-of-00044.safetensors", "model.layers.93.self_attn.q_proj.weight": "model-00031-of-00044.safetensors", "model.layers.86.self_attn.q_proj.weight": "model-00031-of-00044.safetensors", "model.layers.93.self_attn.v_proj.weight": "model-00031-of-00044.safetensors", "model.layers.86.self_attn.v_proj.weight": "model-00031-of-00044.safetensors", "model.layers.94.input_layernorm.weight": "model-00031-of-00044.safetensors", "model.layers.87.input_layernorm.weight": "model-00031-of-00044.safetensors", "model.layers.94.mlp.down_proj.weight": "model-00031-of-00044.safetensors", "model.layers.87.mlp.down_proj.weight": "model-00031-of-00044.safetensors", "model.layers.94.mlp.gate_proj.weight": "model-00031-of-00044.safetensors", "model.layers.87.mlp.gate_proj.weight": "model-00031-of-00044.safetensors", "model.layers.94.mlp.up_proj.weight": "model-00031-of-00044.safetensors", "model.layers.87.mlp.up_proj.weight": "model-00031-of-00044.safetensors", "model.layers.94.post_attention_layernorm.weight": "model-00031-of-00044.safetensors", "model.layers.87.post_attention_layernorm.weight": "model-00031-of-00044.safetensors", "model.layers.94.self_attn.k_proj.weight": "model-00031-of-00044.safetensors", "model.layers.87.self_attn.k_proj.weight": "model-00031-of-00044.safetensors", "model.layers.94.self_attn.o_proj.weight": "model-00031-of-00044.safetensors", "model.layers.87.self_attn.o_proj.weight": "model-00031-of-00044.safetensors", "model.layers.94.self_attn.q_proj.weight": "model-00031-of-00044.safetensors", "model.layers.87.self_attn.q_proj.weight": "model-00031-of-00044.safetensors", "model.layers.94.self_attn.v_proj.weight": "model-00031-of-00044.safetensors", "model.layers.87.self_attn.v_proj.weight": "model-00031-of-00044.safetensors", "model.layers.6.input_layernorm.weight": "model-00031-of-00044.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00032-of-00044.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00032-of-00044.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00032-of-00044.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00032-of-00044.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00032-of-00044.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00032-of-00044.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00032-of-00044.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00032-of-00044.safetensors", "model.layers.95.input_layernorm.weight": "model-00032-of-00044.safetensors", "model.layers.88.input_layernorm.weight": "model-00032-of-00044.safetensors", "model.layers.95.mlp.down_proj.weight": "model-00032-of-00044.safetensors", "model.layers.88.mlp.down_proj.weight": "model-00032-of-00044.safetensors", "model.layers.95.mlp.gate_proj.weight": "model-00032-of-00044.safetensors", "model.layers.88.mlp.gate_proj.weight": "model-00032-of-00044.safetensors", "model.layers.95.mlp.up_proj.weight": "model-00032-of-00044.safetensors", "model.layers.88.mlp.up_proj.weight": "model-00032-of-00044.safetensors", "model.layers.95.post_attention_layernorm.weight": "model-00032-of-00044.safetensors", "model.layers.88.post_attention_layernorm.weight": "model-00032-of-00044.safetensors", "model.layers.95.self_attn.k_proj.weight": "model-00032-of-00044.safetensors", "model.layers.88.self_attn.k_proj.weight": "model-00032-of-00044.safetensors", "model.layers.95.self_attn.o_proj.weight": "model-00032-of-00044.safetensors", "model.layers.88.self_attn.o_proj.weight": "model-00032-of-00044.safetensors", "model.layers.95.self_attn.q_proj.weight": "model-00032-of-00044.safetensors", "model.layers.88.self_attn.q_proj.weight": "model-00033-of-00044.safetensors", "model.layers.95.self_attn.v_proj.weight": "model-00033-of-00044.safetensors", "model.layers.88.self_attn.v_proj.weight": "model-00033-of-00044.safetensors", "model.layers.96.input_layernorm.weight": "model-00033-of-00044.safetensors", "model.layers.89.input_layernorm.weight": "model-00033-of-00044.safetensors", "model.layers.96.mlp.down_proj.weight": "model-00033-of-00044.safetensors", "model.layers.89.mlp.down_proj.weight": "model-00033-of-00044.safetensors", "model.layers.96.mlp.gate_proj.weight": "model-00033-of-00044.safetensors", "model.layers.89.mlp.gate_proj.weight": "model-00033-of-00044.safetensors", "model.layers.96.mlp.up_proj.weight": "model-00033-of-00044.safetensors", "model.layers.89.mlp.up_proj.weight": "model-00033-of-00044.safetensors", "model.layers.96.post_attention_layernorm.weight": "model-00033-of-00044.safetensors", "model.layers.89.post_attention_layernorm.weight": "model-00033-of-00044.safetensors", "model.layers.96.self_attn.k_proj.weight": "model-00033-of-00044.safetensors", "model.layers.89.self_attn.k_proj.weight": "model-00033-of-00044.safetensors", "model.layers.96.self_attn.o_proj.weight": "model-00033-of-00044.safetensors", "model.layers.89.self_attn.o_proj.weight": "model-00033-of-00044.safetensors", "model.layers.96.self_attn.q_proj.weight": "model-00033-of-00044.safetensors", "model.layers.89.self_attn.q_proj.weight": "model-00033-of-00044.safetensors", "model.layers.96.self_attn.v_proj.weight": "model-00033-of-00044.safetensors", "model.layers.89.self_attn.v_proj.weight": "model-00033-of-00044.safetensors", "model.layers.97.input_layernorm.weight": "model-00033-of-00044.safetensors", "model.layers.90.input_layernorm.weight": "model-00033-of-00044.safetensors", "model.layers.97.mlp.down_proj.weight": "model-00033-of-00044.safetensors", "model.layers.90.mlp.down_proj.weight": "model-00033-of-00044.safetensors", "model.layers.97.mlp.gate_proj.weight": "model-00033-of-00044.safetensors", "model.layers.90.mlp.gate_proj.weight": "model-00034-of-00044.safetensors", "model.layers.97.mlp.up_proj.weight": "model-00034-of-00044.safetensors", "model.layers.90.mlp.up_proj.weight": "model-00034-of-00044.safetensors", "model.layers.97.post_attention_layernorm.weight": "model-00034-of-00044.safetensors", "model.layers.90.post_attention_layernorm.weight": "model-00034-of-00044.safetensors", "model.layers.97.self_attn.k_proj.weight": "model-00034-of-00044.safetensors", "model.layers.90.self_attn.k_proj.weight": "model-00034-of-00044.safetensors", "model.layers.97.self_attn.o_proj.weight": "model-00034-of-00044.safetensors", "model.layers.90.self_attn.o_proj.weight": "model-00034-of-00044.safetensors", "model.layers.97.self_attn.q_proj.weight": "model-00034-of-00044.safetensors", "model.layers.90.self_attn.q_proj.weight": "model-00034-of-00044.safetensors", "model.layers.97.self_attn.v_proj.weight": "model-00034-of-00044.safetensors", "model.layers.90.self_attn.v_proj.weight": "model-00034-of-00044.safetensors", "model.layers.98.input_layernorm.weight": "model-00034-of-00044.safetensors", "model.layers.91.input_layernorm.weight": "model-00034-of-00044.safetensors", "model.layers.98.mlp.down_proj.weight": "model-00034-of-00044.safetensors", "model.layers.91.mlp.down_proj.weight": "model-00034-of-00044.safetensors", "model.layers.98.mlp.gate_proj.weight": "model-00034-of-00044.safetensors", "model.layers.91.mlp.gate_proj.weight": "model-00034-of-00044.safetensors", "model.layers.98.mlp.up_proj.weight": "model-00034-of-00044.safetensors", "model.layers.91.mlp.up_proj.weight": "model-00034-of-00044.safetensors", "model.layers.98.post_attention_layernorm.weight": "model-00034-of-00044.safetensors", "model.layers.91.post_attention_layernorm.weight": "model-00034-of-00044.safetensors", "model.layers.98.self_attn.k_proj.weight": "model-00034-of-00044.safetensors", "model.layers.91.self_attn.k_proj.weight": "model-00034-of-00044.safetensors", "model.layers.98.self_attn.o_proj.weight": "model-00034-of-00044.safetensors", "model.layers.91.self_attn.o_proj.weight": "model-00035-of-00044.safetensors", "model.layers.98.self_attn.q_proj.weight": "model-00035-of-00044.safetensors", "model.layers.91.self_attn.q_proj.weight": "model-00035-of-00044.safetensors", "model.layers.98.self_attn.v_proj.weight": "model-00035-of-00044.safetensors", "model.layers.91.self_attn.v_proj.weight": "model-00035-of-00044.safetensors", "model.layers.99.input_layernorm.weight": "model-00035-of-00044.safetensors", "model.layers.99.mlp.down_proj.weight": "model-00035-of-00044.safetensors", "model.layers.99.mlp.gate_proj.weight": "model-00035-of-00044.safetensors", "model.layers.99.mlp.up_proj.weight": "model-00035-of-00044.safetensors", "model.layers.99.post_attention_layernorm.weight": "model-00035-of-00044.safetensors", "model.layers.99.self_attn.k_proj.weight": "model-00035-of-00044.safetensors", "model.layers.99.self_attn.o_proj.weight": "model-00035-of-00044.safetensors", "model.layers.99.self_attn.q_proj.weight": "model-00035-of-00044.safetensors", "model.layers.99.self_attn.v_proj.weight": "model-00035-of-00044.safetensors", "model.layers.107.input_layernorm.weight": "model-00035-of-00044.safetensors", "model.layers.100.input_layernorm.weight": "model-00035-of-00044.safetensors", "model.layers.107.mlp.down_proj.weight": "model-00035-of-00044.safetensors", "model.layers.100.mlp.down_proj.weight": "model-00035-of-00044.safetensors", "model.layers.107.mlp.gate_proj.weight": "model-00035-of-00044.safetensors", "model.layers.100.mlp.gate_proj.weight": "model-00035-of-00044.safetensors", "model.layers.107.mlp.up_proj.weight": "model-00035-of-00044.safetensors", "model.layers.100.mlp.up_proj.weight": "model-00035-of-00044.safetensors", "model.layers.107.post_attention_layernorm.weight": "model-00035-of-00044.safetensors", "model.layers.100.post_attention_layernorm.weight": "model-00035-of-00044.safetensors", "model.layers.107.self_attn.k_proj.weight": "model-00035-of-00044.safetensors", "model.layers.100.self_attn.k_proj.weight": "model-00035-of-00044.safetensors", "model.layers.107.self_attn.o_proj.weight": "model-00036-of-00044.safetensors", "model.layers.100.self_attn.o_proj.weight": "model-00036-of-00044.safetensors", "model.layers.107.self_attn.q_proj.weight": "model-00036-of-00044.safetensors", "model.layers.100.self_attn.q_proj.weight": "model-00036-of-00044.safetensors", "model.layers.107.self_attn.v_proj.weight": "model-00036-of-00044.safetensors", "model.layers.100.self_attn.v_proj.weight": "model-00036-of-00044.safetensors", "model.layers.108.input_layernorm.weight": "model-00036-of-00044.safetensors", "model.layers.101.input_layernorm.weight": "model-00036-of-00044.safetensors", "model.layers.108.mlp.down_proj.weight": "model-00036-of-00044.safetensors", "model.layers.101.mlp.down_proj.weight": "model-00036-of-00044.safetensors", "model.layers.108.mlp.gate_proj.weight": "model-00036-of-00044.safetensors", "model.layers.101.mlp.gate_proj.weight": "model-00036-of-00044.safetensors", "model.layers.108.mlp.up_proj.weight": "model-00036-of-00044.safetensors", "model.layers.101.mlp.up_proj.weight": "model-00036-of-00044.safetensors", "model.layers.108.post_attention_layernorm.weight": "model-00036-of-00044.safetensors", "model.layers.101.post_attention_layernorm.weight": "model-00036-of-00044.safetensors", "model.layers.108.self_attn.k_proj.weight": "model-00036-of-00044.safetensors", "model.layers.101.self_attn.k_proj.weight": "model-00036-of-00044.safetensors", "model.layers.108.self_attn.o_proj.weight": "model-00036-of-00044.safetensors", "model.layers.101.self_attn.o_proj.weight": "model-00036-of-00044.safetensors", "model.layers.108.self_attn.q_proj.weight": "model-00036-of-00044.safetensors", "model.layers.101.self_attn.q_proj.weight": "model-00036-of-00044.safetensors", "model.layers.108.self_attn.v_proj.weight": "model-00036-of-00044.safetensors", "model.layers.101.self_attn.v_proj.weight": "model-00036-of-00044.safetensors", "model.layers.109.input_layernorm.weight": "model-00036-of-00044.safetensors", "model.layers.102.input_layernorm.weight": "model-00036-of-00044.safetensors", "model.layers.109.mlp.down_proj.weight": "model-00036-of-00044.safetensors", "model.layers.102.mlp.down_proj.weight": "model-00036-of-00044.safetensors", "model.layers.109.mlp.gate_proj.weight": "model-00037-of-00044.safetensors", "model.layers.102.mlp.gate_proj.weight": "model-00037-of-00044.safetensors", "model.layers.109.mlp.up_proj.weight": "model-00037-of-00044.safetensors", "model.layers.102.mlp.up_proj.weight": "model-00037-of-00044.safetensors", "model.layers.109.post_attention_layernorm.weight": "model-00037-of-00044.safetensors", "model.layers.102.post_attention_layernorm.weight": "model-00037-of-00044.safetensors", "model.layers.109.self_attn.k_proj.weight": "model-00037-of-00044.safetensors", "model.layers.102.self_attn.k_proj.weight": "model-00037-of-00044.safetensors", "model.layers.109.self_attn.o_proj.weight": "model-00037-of-00044.safetensors", "model.layers.102.self_attn.o_proj.weight": "model-00037-of-00044.safetensors", "model.layers.109.self_attn.q_proj.weight": "model-00037-of-00044.safetensors", "model.layers.102.self_attn.q_proj.weight": "model-00037-of-00044.safetensors", "model.layers.109.self_attn.v_proj.weight": "model-00037-of-00044.safetensors", "model.layers.102.self_attn.v_proj.weight": "model-00037-of-00044.safetensors", "model.layers.110.input_layernorm.weight": "model-00037-of-00044.safetensors", "model.layers.103.input_layernorm.weight": "model-00037-of-00044.safetensors", "model.layers.110.mlp.down_proj.weight": "model-00037-of-00044.safetensors", "model.layers.103.mlp.down_proj.weight": "model-00037-of-00044.safetensors", "model.layers.110.mlp.gate_proj.weight": "model-00037-of-00044.safetensors", "model.layers.103.mlp.gate_proj.weight": "model-00037-of-00044.safetensors", "model.layers.110.mlp.up_proj.weight": "model-00037-of-00044.safetensors", "model.layers.103.mlp.up_proj.weight": "model-00038-of-00044.safetensors", "model.layers.110.post_attention_layernorm.weight": "model-00038-of-00044.safetensors", "model.layers.103.post_attention_layernorm.weight": "model-00038-of-00044.safetensors", "model.layers.110.self_attn.k_proj.weight": "model-00038-of-00044.safetensors", "model.layers.103.self_attn.k_proj.weight": "model-00038-of-00044.safetensors", "model.layers.110.self_attn.o_proj.weight": "model-00038-of-00044.safetensors", "model.layers.103.self_attn.o_proj.weight": "model-00038-of-00044.safetensors", "model.layers.110.self_attn.q_proj.weight": "model-00038-of-00044.safetensors", "model.layers.103.self_attn.q_proj.weight": "model-00038-of-00044.safetensors", "model.layers.110.self_attn.v_proj.weight": "model-00038-of-00044.safetensors", "model.layers.103.self_attn.v_proj.weight": "model-00038-of-00044.safetensors", "model.layers.111.input_layernorm.weight": "model-00038-of-00044.safetensors", "model.layers.104.input_layernorm.weight": "model-00038-of-00044.safetensors", "model.layers.111.mlp.down_proj.weight": "model-00038-of-00044.safetensors", "model.layers.104.mlp.down_proj.weight": "model-00038-of-00044.safetensors", "model.layers.111.mlp.gate_proj.weight": "model-00038-of-00044.safetensors", "model.layers.104.mlp.gate_proj.weight": "model-00038-of-00044.safetensors", "model.layers.111.mlp.up_proj.weight": "model-00038-of-00044.safetensors", "model.layers.104.mlp.up_proj.weight": "model-00038-of-00044.safetensors", "model.layers.111.post_attention_layernorm.weight": "model-00038-of-00044.safetensors", "model.layers.104.post_attention_layernorm.weight": "model-00038-of-00044.safetensors", "model.layers.111.self_attn.k_proj.weight": "model-00038-of-00044.safetensors", "model.layers.104.self_attn.k_proj.weight": "model-00038-of-00044.safetensors", "model.layers.111.self_attn.o_proj.weight": "model-00038-of-00044.safetensors", "model.layers.104.self_attn.o_proj.weight": "model-00038-of-00044.safetensors", "model.layers.111.self_attn.q_proj.weight": "model-00038-of-00044.safetensors", "model.layers.104.self_attn.q_proj.weight": "model-00038-of-00044.safetensors", "model.layers.111.self_attn.v_proj.weight": "model-00038-of-00044.safetensors", "model.layers.104.self_attn.v_proj.weight": "model-00038-of-00044.safetensors", "model.layers.7.input_layernorm.weight": "model-00038-of-00044.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00038-of-00044.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00039-of-00044.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00039-of-00044.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00039-of-00044.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00039-of-00044.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00039-of-00044.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00039-of-00044.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00039-of-00044.safetensors", "model.layers.112.input_layernorm.weight": "model-00039-of-00044.safetensors", "model.layers.105.input_layernorm.weight": "model-00039-of-00044.safetensors", "model.layers.112.mlp.down_proj.weight": "model-00039-of-00044.safetensors", "model.layers.105.mlp.down_proj.weight": "model-00039-of-00044.safetensors", "model.layers.112.mlp.gate_proj.weight": "model-00039-of-00044.safetensors", "model.layers.105.mlp.gate_proj.weight": "model-00039-of-00044.safetensors", "model.layers.112.mlp.up_proj.weight": "model-00039-of-00044.safetensors", "model.layers.105.mlp.up_proj.weight": "model-00039-of-00044.safetensors", "model.layers.112.post_attention_layernorm.weight": "model-00039-of-00044.safetensors", "model.layers.105.post_attention_layernorm.weight": "model-00039-of-00044.safetensors", "model.layers.112.self_attn.k_proj.weight": "model-00039-of-00044.safetensors", "model.layers.105.self_attn.k_proj.weight": "model-00039-of-00044.safetensors", "model.layers.112.self_attn.o_proj.weight": "model-00039-of-00044.safetensors", "model.layers.105.self_attn.o_proj.weight": "model-00039-of-00044.safetensors", "model.layers.112.self_attn.q_proj.weight": "model-00039-of-00044.safetensors", "model.layers.105.self_attn.q_proj.weight": "model-00039-of-00044.safetensors", "model.layers.112.self_attn.v_proj.weight": "model-00039-of-00044.safetensors", "model.layers.105.self_attn.v_proj.weight": "model-00039-of-00044.safetensors", "model.layers.113.input_layernorm.weight": "model-00039-of-00044.safetensors", "model.layers.106.input_layernorm.weight": "model-00039-of-00044.safetensors", "model.layers.113.mlp.down_proj.weight": "model-00040-of-00044.safetensors", "model.layers.106.mlp.down_proj.weight": "model-00040-of-00044.safetensors", "model.layers.113.mlp.gate_proj.weight": "model-00040-of-00044.safetensors", "model.layers.106.mlp.gate_proj.weight": "model-00040-of-00044.safetensors", "model.layers.113.mlp.up_proj.weight": "model-00040-of-00044.safetensors", "model.layers.106.mlp.up_proj.weight": "model-00040-of-00044.safetensors", "model.layers.113.post_attention_layernorm.weight": "model-00040-of-00044.safetensors", "model.layers.106.post_attention_layernorm.weight": "model-00040-of-00044.safetensors", "model.layers.113.self_attn.k_proj.weight": "model-00040-of-00044.safetensors", "model.layers.106.self_attn.k_proj.weight": "model-00040-of-00044.safetensors", "model.layers.113.self_attn.o_proj.weight": "model-00040-of-00044.safetensors", "model.layers.106.self_attn.o_proj.weight": "model-00040-of-00044.safetensors", "model.layers.113.self_attn.q_proj.weight": "model-00040-of-00044.safetensors", "model.layers.106.self_attn.q_proj.weight": "model-00040-of-00044.safetensors", "model.layers.113.self_attn.v_proj.weight": "model-00040-of-00044.safetensors", "model.layers.106.self_attn.v_proj.weight": "model-00040-of-00044.safetensors", "model.layers.114.input_layernorm.weight": "model-00040-of-00044.safetensors", "model.layers.114.mlp.down_proj.weight": "model-00040-of-00044.safetensors", "model.layers.114.mlp.gate_proj.weight": "model-00040-of-00044.safetensors", "model.layers.114.mlp.up_proj.weight": "model-00040-of-00044.safetensors", "model.layers.114.post_attention_layernorm.weight": "model-00040-of-00044.safetensors", "model.layers.114.self_attn.k_proj.weight": "model-00040-of-00044.safetensors", "model.layers.114.self_attn.o_proj.weight": "model-00040-of-00044.safetensors", "model.layers.114.self_attn.q_proj.weight": "model-00041-of-00044.safetensors", "model.layers.114.self_attn.v_proj.weight": "model-00041-of-00044.safetensors", "model.layers.115.input_layernorm.weight": "model-00041-of-00044.safetensors", "model.layers.115.mlp.down_proj.weight": "model-00041-of-00044.safetensors", "model.layers.115.mlp.gate_proj.weight": "model-00041-of-00044.safetensors", "model.layers.115.mlp.up_proj.weight": "model-00041-of-00044.safetensors", "model.layers.115.post_attention_layernorm.weight": "model-00041-of-00044.safetensors", "model.layers.115.self_attn.k_proj.weight": "model-00041-of-00044.safetensors", "model.layers.115.self_attn.o_proj.weight": "model-00041-of-00044.safetensors", "model.layers.115.self_attn.q_proj.weight": "model-00041-of-00044.safetensors", "model.layers.115.self_attn.v_proj.weight": "model-00041-of-00044.safetensors", "model.layers.116.input_layernorm.weight": "model-00041-of-00044.safetensors", "model.layers.116.mlp.down_proj.weight": "model-00041-of-00044.safetensors", "model.layers.116.mlp.gate_proj.weight": "model-00041-of-00044.safetensors", "model.layers.116.mlp.up_proj.weight": "model-00041-of-00044.safetensors", "model.layers.116.post_attention_layernorm.weight": "model-00041-of-00044.safetensors", "model.layers.116.self_attn.k_proj.weight": "model-00041-of-00044.safetensors", "model.layers.116.self_attn.o_proj.weight": "model-00041-of-00044.safetensors", "model.layers.116.self_attn.q_proj.weight": "model-00041-of-00044.safetensors", "model.layers.116.self_attn.v_proj.weight": "model-00041-of-00044.safetensors", "model.layers.117.input_layernorm.weight": "model-00041-of-00044.safetensors", "model.layers.117.mlp.down_proj.weight": "model-00041-of-00044.safetensors", "model.layers.117.mlp.gate_proj.weight": "model-00041-of-00044.safetensors", "model.layers.117.mlp.up_proj.weight": "model-00041-of-00044.safetensors", "model.layers.117.post_attention_layernorm.weight": "model-00041-of-00044.safetensors", "model.layers.117.self_attn.k_proj.weight": "model-00041-of-00044.safetensors", "model.layers.117.self_attn.o_proj.weight": "model-00042-of-00044.safetensors", "model.layers.117.self_attn.q_proj.weight": "model-00042-of-00044.safetensors", "model.layers.117.self_attn.v_proj.weight": "model-00042-of-00044.safetensors", "model.layers.118.input_layernorm.weight": "model-00042-of-00044.safetensors", "model.layers.118.mlp.down_proj.weight": "model-00042-of-00044.safetensors", "model.layers.118.mlp.gate_proj.weight": "model-00042-of-00044.safetensors", "model.layers.118.mlp.up_proj.weight": "model-00042-of-00044.safetensors", "model.layers.118.post_attention_layernorm.weight": "model-00042-of-00044.safetensors", "model.layers.118.self_attn.k_proj.weight": "model-00042-of-00044.safetensors", "model.layers.118.self_attn.o_proj.weight": "model-00042-of-00044.safetensors", "model.layers.118.self_attn.q_proj.weight": "model-00042-of-00044.safetensors", "model.layers.118.self_attn.v_proj.weight": "model-00042-of-00044.safetensors", "model.layers.119.input_layernorm.weight": "model-00042-of-00044.safetensors", "model.layers.119.mlp.down_proj.weight": "model-00042-of-00044.safetensors", "model.layers.119.mlp.gate_proj.weight": "model-00042-of-00044.safetensors", "model.layers.119.mlp.up_proj.weight": "model-00042-of-00044.safetensors", "model.layers.119.post_attention_layernorm.weight": "model-00042-of-00044.safetensors", "model.layers.119.self_attn.k_proj.weight": "model-00042-of-00044.safetensors", "model.layers.119.self_attn.o_proj.weight": "model-00042-of-00044.safetensors", "model.layers.119.self_attn.q_proj.weight": "model-00042-of-00044.safetensors", "model.layers.119.self_attn.v_proj.weight": "model-00042-of-00044.safetensors", "model.layers.120.input_layernorm.weight": "model-00042-of-00044.safetensors", "model.layers.120.mlp.down_proj.weight": "model-00042-of-00044.safetensors", "model.layers.120.mlp.gate_proj.weight": "model-00042-of-00044.safetensors", "model.layers.120.mlp.up_proj.weight": "model-00043-of-00044.safetensors", "model.layers.120.post_attention_layernorm.weight": "model-00043-of-00044.safetensors", "model.layers.120.self_attn.k_proj.weight": "model-00043-of-00044.safetensors", "model.layers.120.self_attn.o_proj.weight": "model-00043-of-00044.safetensors", "model.layers.120.self_attn.q_proj.weight": "model-00043-of-00044.safetensors", "model.layers.120.self_attn.v_proj.weight": "model-00043-of-00044.safetensors", "model.layers.121.input_layernorm.weight": "model-00043-of-00044.safetensors", "model.layers.121.mlp.down_proj.weight": "model-00043-of-00044.safetensors", "model.layers.121.mlp.gate_proj.weight": "model-00043-of-00044.safetensors", "model.layers.121.mlp.up_proj.weight": "model-00043-of-00044.safetensors", "model.layers.121.post_attention_layernorm.weight": "model-00043-of-00044.safetensors", "model.layers.121.self_attn.k_proj.weight": "model-00043-of-00044.safetensors", "model.layers.121.self_attn.o_proj.weight": "model-00043-of-00044.safetensors", "model.layers.121.self_attn.q_proj.weight": "model-00043-of-00044.safetensors", "model.layers.121.self_attn.v_proj.weight": "model-00043-of-00044.safetensors", "model.layers.16.input_layernorm.weight": "model-00043-of-00044.safetensors", "model.layers.8.input_layernorm.weight": "model-00043-of-00044.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00043-of-00044.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00043-of-00044.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00043-of-00044.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00043-of-00044.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00043-of-00044.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00044-of-00044.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00044-of-00044.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00044-of-00044.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00044-of-00044.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00044-of-00044.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00044-of-00044.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00044-of-00044.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00044-of-00044.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00044-of-00044.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00044-of-00044.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00044-of-00044.safetensors", "model.layers.17.input_layernorm.weight": "model-00044-of-00044.safetensors", "model.layers.9.input_layernorm.weight": "model-00044-of-00044.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00044-of-00044.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00044-of-00044.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00044-of-00044.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00044-of-00044.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00044-of-00044.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00044-of-00044.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00044-of-00044.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00044-of-00044.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00044-of-00044.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00044-of-00044.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00044-of-00044.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00044-of-00044.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00044-of-00044.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00044-of-00044.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00044-of-00044.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00044-of-00044.safetensors", "model.norm.weight": "model-00044-of-00044.safetensors"}}

special_tokens_map.json

@@ -0,0 +1,16 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|end_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}