Upload folder using huggingface_hub

README.md

@@ -0,0 +1,48 @@
+---
+base_model:
+- NousResearch/Meta-Llama-3.1-70B
+- NousResearch/Llama-2-70b-hf
+- NousResearch/Hermes-2-Pro-Llama-3-70B
+library_name: transformers
+tags:
+- mergekit
+- merge
+
+---
+# merge
+
+This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
+
+## Merge Details
+### Merge Method
+
+This model was merged using the [TIES](https://arxiv.org/abs/2306.01708) merge method using [NousResearch/Llama-2-70b-hf](https://huggingface.co/NousResearch/Llama-2-70b-hf) as a base.
+
+### Models Merged
+
+The following models were included in the merge:
+* [NousResearch/Meta-Llama-3.1-70B](https://huggingface.co/NousResearch/Meta-Llama-3.1-70B)
+* [NousResearch/Hermes-2-Pro-Llama-3-70B](https://huggingface.co/NousResearch/Hermes-2-Pro-Llama-3-70B)
+
+### Configuration
+
+The following YAML configuration was used to produce this model:
+
+```yaml
+models:
+  - model: NousResearch/Meta-Llama-3.1-70B
+    parameters:
+      density: 0.5
+      weight: 0.5
+  - model: NousResearch/Hermes-2-Pro-Llama-3-70B
+    parameters:
+      density: 0.5
+      weight: 0.5
+
+merge_method: ties
+base_model: NousResearch/Llama-2-70b-hf
+parameters:
+  normalize: false
+  int8_mask: true
+dtype: float16
+```

config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "NousResearch/Llama-2-70b-hf",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 28672,
+  "max_position_embeddings": 4096,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 80,
+  "num_key_value_heads": 8,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.44.1",
+  "use_cache": true,
+  "vocab_size": 32000
+}

mergekit_config.yml

@@ -0,0 +1,16 @@
+models:
+  - model: NousResearch/Meta-Llama-3.1-70B
+    parameters:
+      density: 0.5
+      weight: 0.5
+  - model: NousResearch/Hermes-2-Pro-Llama-3-70B
+    parameters:
+      density: 0.5
+      weight: 0.5
+
+merge_method: ties
+base_model: NousResearch/Llama-2-70b-hf
+parameters:
+  normalize: false
+  int8_mask: true
+dtype: float16

model-00001-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b57f4147ee06cc620c09c55426d82aa62a83d5c56e9e796585bb44e3ce752d0
+size 4940974552

model-00002-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86f10ee8c44a2cf3f2b6a0daa9f947c37fd54e9a21689e5823c44a10026c2ebd
+size 4664167392

model-00003-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:441727ec731dff0954c4062a771073ff2dea3f88a5fd3edeb4a68f0e09be913d
+size 4982917880

model-00004-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34b3eb486f53af879c1d89ee4a061c613c2727e2ec22883f3ffa3dc00e80655c
+size 4999711704

model-00005-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3b35f5a1724ab88bc39eb0296f82a568ec14fe0ea38a8a5eb0fa3a9b9d56189
+size 4647373544

model-00006-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6248ade824523678941bb960327f5b895c9a2027356798d2f65b4421bc030e7d
+size 4664167384

model-00007-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb13af442474abb92bcc383d7e4a65e51cb930352db22179f77a52051e91bf88
+size 4664167392

model-00008-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85f94d358c2f32581e93ae8387ab1deced487d07cb47e1c2afbb18eb2eb97312
+size 4982917880

model-00009-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9917029e0f4d271742e099a5aa0e814918b7600f3fec748f39ed939192cad20
+size 4999711696

model-00010-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:053cbfc81d8f26f692c651d03f0acee8f78f8aa2e14d3a092801df67d64f7bf5
+size 4647373552

model-00011-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99cf29fc2ae9ecfef537da56963f1da9d43ab60192db7c02802cf235024e2977
+size 4664167384

model-00012-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32877a486d85b01385adab222bddcc13865ab4cc1eea22d3d248f20ed9a76778
+size 4664167392

model-00013-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5030598cc22c84ea3c063af03ff507b1bdfb2238725760d72ef436bac65a625b
+size 4982917872

model-00014-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55a579875189944e16ffaaf3d0e1d2ce518611c4eead3b8c56117cd910dfebe8
+size 4999711704

model-00015-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb32c5bc98b4fdb0bb7939a7093e38c3e449cfb7e6d18e554d9fad21daf89094
+size 4647373552

model-00016-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a98880348cfdae80b94108ed2d6a9b2aff889ef41b80d495f72eb77567eda4fc
+size 4664167384

model-00017-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d3c99b779cbabb0239621f19001149ec6d85be7f5553c13cde6ea2ae2ff8d31
+size 4664167384

model-00018-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98461454211ef03fe7b88573757bbe276a4404868ea6dc994e3159f8c84b3c38
+size 4982917880

model-00019-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bf809c336cfd8bd85849be6ca082e88e816d1cdcc58c69bf93aed0d2f17122b
+size 4999711704

model-00020-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:218eb52bafbd81e5fc08c9efec9068933bd52f4ed5c449eb355f1c17ed75c3a9
+size 4647373552

model-00021-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05262fb655b187c619056531b17779a2680c37274c1418f905eb6756f4e6687c
+size 4664167376

model-00022-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0b1d590c141b6cdb4d4ca7d0eb6cfffb8ab399b994d06b770fcb196f19cedeb
+size 4664167392

model-00023-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f1c1f24bf9dfb24ebce05931eb57c2a20f2de7d9b298d8a508bbebe1980297e
+size 4982917880

model-00024-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6226b38380f6cec4a835a4bb69a5e1cfbb091b1b7e3ad30f8aa1098527453e18
+size 4999711704

model-00025-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c789c1eb4c6fa79ca75b99465a628d46e32eeb0d8bae3a6d6182120bdabe776
+size 4647373544

model-00026-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:737d9f3624f7bf93628d36f0c799171d0408c7392c1c9204e8ae2dea503ef9eb
+size 4664167384

model-00027-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f1d4a055a3367bcff294f1e0db0fba952e71cdd276308147082e55b6fe73f23
+size 4664167392

model-00028-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fefa7b95a860ba64c60c4e67b498e2a12e6e44ab2fc8fcd3945269fb6a5532f1
+size 4982917880

model-00029-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12fead8865ed5a8e850ebdfb1440b23a017441bcfb65a6cc9fc4373c82447d43
+size 3573631344

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.4.4", "total_size": 137953296384}, "weight_map": {"lm_head.weight": "model-00001-of-00029.safetensors", "model.embed_tokens.weight": "model-00001-of-00029.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00029.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00029.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00029.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00029.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00029.safetensors", "model.layers.10.input_layernorm.weight": "model-00001-of-00029.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00001-of-00029.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00029.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00002-of-00029.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00029.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00029.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00029.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.input_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.input_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00029.safetensors", "model.layers.13.input_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00003-of-00029.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.input_layernorm.weight": "model-00003-of-00029.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00003-of-00029.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.input_layernorm.weight": "model-00003-of-00029.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00003-of-00029.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00029.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.input_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.input_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00029.safetensors", "model.layers.18.input_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00004-of-00029.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00004-of-00029.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00004-of-00029.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00004-of-00029.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00029.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00029.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.input_layernorm.weight": "model-00005-of-00029.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00029.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.input_layernorm.weight": "model-00005-of-00029.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00005-of-00029.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00005-of-00029.safetensors", "model.layers.20.input_layernorm.weight": "model-00005-of-00029.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00005-of-00029.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00029.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00006-of-00029.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00006-of-00029.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00006-of-00029.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00006-of-00029.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.input_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.input_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00006-of-00029.safetensors", "model.layers.23.input_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00006-of-00029.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00007-of-00029.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00007-of-00029.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00007-of-00029.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00007-of-00029.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00007-of-00029.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.input_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.input_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00007-of-00029.safetensors", "model.layers.26.input_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00008-of-00029.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.input_layernorm.weight": "model-00008-of-00029.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00008-of-00029.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.input_layernorm.weight": "model-00008-of-00029.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00008-of-00029.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00009-of-00029.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.input_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.input_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00009-of-00029.safetensors", "model.layers.30.input_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00009-of-00029.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00009-of-00029.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00009-of-00029.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00009-of-00029.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00010-of-00029.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00010-of-00029.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.input_layernorm.weight": "model-00010-of-00029.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00010-of-00029.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.input_layernorm.weight": "model-00010-of-00029.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00010-of-00029.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00010-of-00029.safetensors", "model.layers.33.input_layernorm.weight": "model-00010-of-00029.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00010-of-00029.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00010-of-00029.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00011-of-00029.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00011-of-00029.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00011-of-00029.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00011-of-00029.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.input_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.input_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00011-of-00029.safetensors", "model.layers.36.input_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00011-of-00029.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00012-of-00029.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00012-of-00029.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00012-of-00029.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00012-of-00029.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00012-of-00029.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.input_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.input_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00012-of-00029.safetensors", "model.layers.39.input_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00013-of-00029.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.input_layernorm.weight": "model-00013-of-00029.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00013-of-00029.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.input_layernorm.weight": "model-00013-of-00029.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00013-of-00029.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00014-of-00029.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.input_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.input_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00014-of-00029.safetensors", "model.layers.43.input_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00014-of-00029.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00014-of-00029.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00014-of-00029.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00014-of-00029.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00015-of-00029.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00015-of-00029.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.input_layernorm.weight": "model-00015-of-00029.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00015-of-00029.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.input_layernorm.weight": "model-00015-of-00029.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00015-of-00029.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00015-of-00029.safetensors", "model.layers.46.input_layernorm.weight": "model-00015-of-00029.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00015-of-00029.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00015-of-00029.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00016-of-00029.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00016-of-00029.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00016-of-00029.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00016-of-00029.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.input_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.input_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00016-of-00029.safetensors", "model.layers.49.input_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00016-of-00029.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00017-of-00029.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00017-of-00029.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00017-of-00029.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00017-of-00029.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00017-of-00029.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.input_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.input_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00017-of-00029.safetensors", "model.layers.51.input_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00018-of-00029.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.input_layernorm.weight": "model-00018-of-00029.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00018-of-00029.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.input_layernorm.weight": "model-00018-of-00029.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00018-of-00029.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00019-of-00029.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.input_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.input_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00019-of-00029.safetensors", "model.layers.56.input_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00019-of-00029.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00019-of-00029.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00019-of-00029.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00019-of-00029.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00020-of-00029.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00020-of-00029.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.input_layernorm.weight": "model-00020-of-00029.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00020-of-00029.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.input_layernorm.weight": "model-00020-of-00029.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00020-of-00029.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00020-of-00029.safetensors", "model.layers.59.input_layernorm.weight": "model-00020-of-00029.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00020-of-00029.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00020-of-00029.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00021-of-00029.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00021-of-00029.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00021-of-00029.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00021-of-00029.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.input_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.input_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00021-of-00029.safetensors", "model.layers.61.input_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00021-of-00029.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00022-of-00029.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00022-of-00029.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00022-of-00029.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00022-of-00029.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00022-of-00029.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.input_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.input_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00022-of-00029.safetensors", "model.layers.64.input_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00023-of-00029.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.input_layernorm.weight": "model-00023-of-00029.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00023-of-00029.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.input_layernorm.weight": "model-00023-of-00029.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00023-of-00029.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00024-of-00029.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.input_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.input_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00024-of-00029.safetensors", "model.layers.69.input_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00024-of-00029.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00024-of-00029.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00024-of-00029.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00024-of-00029.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00025-of-00029.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00025-of-00029.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.input_layernorm.weight": "model-00025-of-00029.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00025-of-00029.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.input_layernorm.weight": "model-00025-of-00029.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00025-of-00029.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00025-of-00029.safetensors", "model.layers.71.input_layernorm.weight": "model-00025-of-00029.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00025-of-00029.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00025-of-00029.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00026-of-00029.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00026-of-00029.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00026-of-00029.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00026-of-00029.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.input_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.input_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00026-of-00029.safetensors", "model.layers.74.input_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00026-of-00029.safetensors", "model.layers.74.mlp.gate_proj.weight": "model-00027-of-00029.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00027-of-00029.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00027-of-00029.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00027-of-00029.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00027-of-00029.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.input_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.input_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00027-of-00029.safetensors", "model.layers.77.input_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00028-of-00029.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.input_layernorm.weight": "model-00028-of-00029.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00028-of-00029.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.input_layernorm.weight": "model-00028-of-00029.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00028-of-00029.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00029-of-00029.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.input_layernorm.weight": "model-00029-of-00029.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00029-of-00029.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.input_layernorm.weight": "model-00029-of-00029.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00029-of-00029.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00029-of-00029.safetensors", "model.norm.weight": "model-00029-of-00029.safetensors"}}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

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

tokenizer_config.json

@@ -0,0 +1,42 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<unk>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}