Upload folder using huggingface_hub

README.md

@@ -0,0 +1,58 @@
+---
+base_model:
+- unsloth/Mistral-Small-Instruct-2409
+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 passthrough merge method.
+
+### Models Merged
+
+The following models were included in the merge:
+* [unsloth/Mistral-Small-Instruct-2409](https://huggingface.co/unsloth/Mistral-Small-Instruct-2409)
+
+### Configuration
+
+The following YAML configuration was used to produce this model:
+
+```yaml
+dtype: bfloat16
+merge_method: passthrough
+slices:
+- sources:
+  - layer_range: [0, 48]
+    model: unsloth/Mistral-Small-Instruct-2409
+- sources:
+  - layer_range: [14, 48]
+    model: unsloth/Mistral-Small-Instruct-2409
+    parameters:
+      scale:
+      - filter: o_proj
+        value: 0.0
+      - filter: down_proj
+        value: 0.0
+      - value: 1.0
+- sources:
+  - layer_range: [14, 48]
+    model: unsloth/Mistral-Small-Instruct-2409
+    parameters:
+      scale:
+      - filter: o_proj
+        value: 0.0
+      - filter: down_proj
+        value: 0.0
+      - value: 1.0
+- sources:
+  - layer_range: [48, 55]
+    model: unsloth/Mistral-Small-Instruct-2409
+```

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "unsloth/Mistral-Small-Instruct-2409",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 6144,
+  "initializer_range": 0.02,
+  "intermediate_size": 16384,
+  "max_position_embeddings": 131072,
+  "model_type": "mistral",
+  "num_attention_heads": 48,
+  "num_hidden_layers": 123,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.43.2",
+  "use_cache": true,
+  "vocab_size": 32768
+}

mergekit_config.yml

@@ -0,0 +1,29 @@
+dtype: bfloat16
+merge_method: passthrough
+slices:
+- sources:
+  - layer_range: [0, 48]
+    model: unsloth/Mistral-Small-Instruct-2409
+- sources:
+  - layer_range: [14, 48]
+    model: unsloth/Mistral-Small-Instruct-2409
+    parameters:
+      scale:
+      - filter: o_proj
+        value: 0.0
+      - filter: down_proj
+        value: 0.0
+      - value: 1.0
+- sources:
+  - layer_range: [14, 48]
+    model: unsloth/Mistral-Small-Instruct-2409
+    parameters:
+      scale:
+      - filter: o_proj
+        value: 0.0
+      - filter: down_proj
+        value: 0.0
+      - value: 1.0
+- sources:
+  - layer_range: [48, 55]
+    model: unsloth/Mistral-Small-Instruct-2409

model-00001-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ff50d62ecaf4826b998e558aee7a30881d340a018dd012ceddc5dc7d558a1fe
+size 4907476544

model-00002-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca6924506b3d1cad8386f58e4319cd48d89361f6c9e81a28e5c2aaa56c7d7352
+size 4995582240

model-00003-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af8902c0f138c6e76764d963e92eaf00ddd53c2df1ab6d0b8288eea725ef3512
+size 4945276088

model-00004-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f89ec7f9d4dfdd6608b42f2e778a8e28746ea75070c958424006e35b1cc6ac4d
+size 4882298944

model-00005-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:465ca4cec8605dedb7c5bc4d3d74d6a06c8687a1b8b62292d285f25545c44364
+size 4995582240

model-00006-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d121af8f4f419ae8ff58511713e6b443744c5af10a6fb854457b873955ead9a9
+size 4945276088

model-00007-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:499a2a650da2e03357b069c91fceadbb0747f44bdb5c52a9593350e4c96e4073
+size 4882323752

model-00008-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55987d11b3aeb3918c051413b62425a02cc959e03f54dc9d3022c0182b363bb9
+size 4882323752

model-00009-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f1c26a5a2c5820b5cb31e528b46a89a3afee17a8786526ea61d514bf8d9b8d0
+size 4970391968

model-00010-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2ac6c4868d2677ad21dea7fcbd6dcecdd323db534c604c90247b0a59be33bcf
+size 4970441560

model-00011-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7985aaf780fb8fba98db0f818f17fd0949cd2379852ea06b092aead06d85c196
+size 4882323768

model-00012-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:304e6f4875d70c3261d65e46f0e0cca26f56a531acb66f37a063653f736d0d32
+size 4882323768

model-00013-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ee9e4815d5ae35b2b3f19cb88e1d9f3e48edcb2956713711a6de8216c4baf06
+size 4882323768

model-00014-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9675aa1ddd1b992912ac8c142b2fb2c5432f7e7145d1cb48832bcf472ac07e64
+size 4857183088

model-00015-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4601dc53979e1ff005221dbfef380bb82514e06b075a4ad699b849aec7525fbe
+size 4882323768

model-00016-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5567dba6581b06f4caa3fc71d202f82433d1eba35a91c1584f3338cc9ec27ca
+size 4882323768

model-00017-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c098ee49c8943d9940a625f6752e317ce8b74418a40c15a55e344186c1273ab
+size 4882323768

model-00018-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9023bfb6c43490b259cab73dfbf07a31163e13e8494fe4632654c8029fa57fb1
+size 4857158312

model-00019-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:012d9850923ea8435e4614b62cd34f7df76c09c10cbef7e4b7805f54c426ac3c
+size 4882323784

model-00020-of-00020.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34551f3902ae78d1b62ea45a8d84946ea94b25970d6a0ea76c97e8780da79f4b
+size 3498177400

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.4.4", "total_size": 96765628416}, "weight_map": {"lm_head.weight": "model-00001-of-00020.safetensors", "model.embed_tokens.weight": "model-00001-of-00020.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00020.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00001-of-00020.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00020.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00001-of-00020.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00020.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00020.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00020.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00020.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00020.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00020.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00001-of-00020.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00020.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00001-of-00020.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00020.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00020.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00020.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00020.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00020.safetensors", "model.layers.10.input_layernorm.weight": "model-00001-of-00020.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00001-of-00020.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00020.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00001-of-00020.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00020.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00020.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00020.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00020.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00020.safetensors", "model.layers.11.input_layernorm.weight": "model-00001-of-00020.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00001-of-00020.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00020.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00001-of-00020.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00020.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00020.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00020.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00020.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00020.safetensors", "model.layers.12.input_layernorm.weight": "model-00001-of-00020.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00001-of-00020.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00020.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00001-of-00020.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00020.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00020.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00020.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00020.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00020.safetensors", "model.layers.13.input_layernorm.weight": "model-00001-of-00020.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00001-of-00020.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00020.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00002-of-00020.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00020.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00020.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00020.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00020.safetensors", "model.layers.82.input_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.48.input_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.14.input_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.82.mlp.down_proj.weight": "model-00002-of-00020.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00002-of-00020.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00002-of-00020.safetensors", "model.layers.82.mlp.gate_proj.weight": "model-00002-of-00020.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00002-of-00020.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00020.safetensors", "model.layers.82.mlp.up_proj.weight": "model-00002-of-00020.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00002-of-00020.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00002-of-00020.safetensors", "model.layers.82.post_attention_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.82.self_attn.k_proj.weight": "model-00002-of-00020.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00002-of-00020.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00020.safetensors", "model.layers.82.self_attn.o_proj.weight": "model-00002-of-00020.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00002-of-00020.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00020.safetensors", "model.layers.82.self_attn.q_proj.weight": "model-00002-of-00020.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00002-of-00020.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00020.safetensors", "model.layers.82.self_attn.v_proj.weight": "model-00002-of-00020.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00002-of-00020.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00020.safetensors", "model.layers.83.input_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.49.input_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.15.input_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.83.mlp.down_proj.weight": "model-00002-of-00020.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00002-of-00020.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00002-of-00020.safetensors", "model.layers.83.mlp.gate_proj.weight": "model-00002-of-00020.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00002-of-00020.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00020.safetensors", "model.layers.83.mlp.up_proj.weight": "model-00002-of-00020.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00002-of-00020.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00002-of-00020.safetensors", "model.layers.83.post_attention_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00020.safetensors", "model.layers.83.self_attn.k_proj.weight": "model-00002-of-00020.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00002-of-00020.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00020.safetensors", "model.layers.83.self_attn.o_proj.weight": "model-00002-of-00020.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00002-of-00020.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00020.safetensors", "model.layers.83.self_attn.q_proj.weight": "model-00003-of-00020.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00003-of-00020.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00003-of-00020.safetensors", "model.layers.83.self_attn.v_proj.weight": "model-00003-of-00020.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00003-of-00020.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00003-of-00020.safetensors", "model.layers.84.input_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.50.input_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.16.input_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.84.mlp.down_proj.weight": "model-00003-of-00020.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00003-of-00020.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00003-of-00020.safetensors", "model.layers.84.mlp.gate_proj.weight": "model-00003-of-00020.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00003-of-00020.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00003-of-00020.safetensors", "model.layers.84.mlp.up_proj.weight": "model-00003-of-00020.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00003-of-00020.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00003-of-00020.safetensors", "model.layers.84.post_attention_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.84.self_attn.k_proj.weight": "model-00003-of-00020.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00003-of-00020.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00003-of-00020.safetensors", "model.layers.84.self_attn.o_proj.weight": "model-00003-of-00020.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00003-of-00020.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00003-of-00020.safetensors", "model.layers.84.self_attn.q_proj.weight": "model-00003-of-00020.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00003-of-00020.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00003-of-00020.safetensors", "model.layers.84.self_attn.v_proj.weight": "model-00003-of-00020.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00003-of-00020.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00003-of-00020.safetensors", "model.layers.85.input_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.51.input_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.17.input_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.85.mlp.down_proj.weight": "model-00003-of-00020.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00003-of-00020.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00003-of-00020.safetensors", "model.layers.85.mlp.gate_proj.weight": "model-00003-of-00020.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00003-of-00020.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00003-of-00020.safetensors", "model.layers.85.mlp.up_proj.weight": "model-00003-of-00020.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00003-of-00020.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00003-of-00020.safetensors", "model.layers.85.post_attention_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.85.self_attn.k_proj.weight": "model-00003-of-00020.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00003-of-00020.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00003-of-00020.safetensors", "model.layers.85.self_attn.o_proj.weight": "model-00003-of-00020.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00003-of-00020.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00003-of-00020.safetensors", "model.layers.85.self_attn.q_proj.weight": "model-00003-of-00020.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00003-of-00020.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00003-of-00020.safetensors", "model.layers.85.self_attn.v_proj.weight": "model-00003-of-00020.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00003-of-00020.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00003-of-00020.safetensors", "model.layers.86.input_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.52.input_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.18.input_layernorm.weight": "model-00003-of-00020.safetensors", "model.layers.86.mlp.down_proj.weight": "model-00004-of-00020.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00004-of-00020.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00004-of-00020.safetensors", "model.layers.86.mlp.gate_proj.weight": "model-00004-of-00020.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00004-of-00020.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00004-of-00020.safetensors", "model.layers.86.mlp.up_proj.weight": "model-00004-of-00020.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00004-of-00020.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00004-of-00020.safetensors", "model.layers.86.post_attention_layernorm.weight": "model-00004-of-00020.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00004-of-00020.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00004-of-00020.safetensors", "model.layers.86.self_attn.k_proj.weight": "model-00004-of-00020.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00004-of-00020.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00004-of-00020.safetensors", "model.layers.86.self_attn.o_proj.weight": "model-00004-of-00020.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00004-of-00020.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00004-of-00020.safetensors", "model.layers.86.self_attn.q_proj.weight": "model-00004-of-00020.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00004-of-00020.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00004-of-00020.safetensors", "model.layers.86.self_attn.v_proj.weight": "model-00004-of-00020.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00004-of-00020.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00004-of-00020.safetensors", "model.layers.87.input_layernorm.weight": "model-00004-of-00020.safetensors", "model.layers.53.input_layernorm.weight": "model-00004-of-00020.safetensors", "model.layers.19.input_layernorm.weight": "model-00004-of-00020.safetensors", "model.layers.87.mlp.down_proj.weight": "model-00004-of-00020.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00004-of-00020.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00004-of-00020.safetensors", "model.layers.87.mlp.gate_proj.weight": "model-00004-of-00020.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00004-of-00020.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00004-of-00020.safetensors", "model.layers.87.mlp.up_proj.weight": "model-00004-of-00020.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00004-of-00020.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00004-of-00020.safetensors", "model.layers.87.post_attention_layernorm.weight": "model-00004-of-00020.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00004-of-00020.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00004-of-00020.safetensors", "model.layers.87.self_attn.k_proj.weight": "model-00004-of-00020.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00004-of-00020.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00004-of-00020.safetensors", "model.layers.87.self_attn.o_proj.weight": "model-00004-of-00020.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00004-of-00020.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00004-of-00020.safetensors", "model.layers.87.self_attn.q_proj.weight": "model-00004-of-00020.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00004-of-00020.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00004-of-00020.safetensors", "model.layers.87.self_attn.v_proj.weight": "model-00004-of-00020.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00004-of-00020.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00004-of-00020.safetensors", "model.layers.2.input_layernorm.weight": "model-00004-of-00020.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00004-of-00020.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00005-of-00020.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00005-of-00020.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00005-of-00020.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00005-of-00020.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00005-of-00020.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00005-of-00020.safetensors", "model.layers.88.input_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.54.input_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.20.input_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.88.mlp.down_proj.weight": "model-00005-of-00020.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00005-of-00020.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00005-of-00020.safetensors", "model.layers.88.mlp.gate_proj.weight": "model-00005-of-00020.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00005-of-00020.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00020.safetensors", "model.layers.88.mlp.up_proj.weight": "model-00005-of-00020.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00005-of-00020.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00005-of-00020.safetensors", "model.layers.88.post_attention_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.88.self_attn.k_proj.weight": "model-00005-of-00020.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00005-of-00020.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00020.safetensors", "model.layers.88.self_attn.o_proj.weight": "model-00005-of-00020.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00005-of-00020.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00020.safetensors", "model.layers.88.self_attn.q_proj.weight": "model-00005-of-00020.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00005-of-00020.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00020.safetensors", "model.layers.88.self_attn.v_proj.weight": "model-00005-of-00020.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00005-of-00020.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00020.safetensors", "model.layers.89.input_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.55.input_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.21.input_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.89.mlp.down_proj.weight": "model-00005-of-00020.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00005-of-00020.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00005-of-00020.safetensors", "model.layers.89.mlp.gate_proj.weight": "model-00005-of-00020.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00005-of-00020.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00020.safetensors", "model.layers.89.mlp.up_proj.weight": "model-00005-of-00020.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00005-of-00020.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00005-of-00020.safetensors", "model.layers.89.post_attention_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00020.safetensors", "model.layers.89.self_attn.k_proj.weight": "model-00005-of-00020.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00005-of-00020.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00020.safetensors", "model.layers.89.self_attn.o_proj.weight": "model-00005-of-00020.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00005-of-00020.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00020.safetensors", "model.layers.89.self_attn.q_proj.weight": "model-00006-of-00020.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00006-of-00020.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00006-of-00020.safetensors", "model.layers.89.self_attn.v_proj.weight": "model-00006-of-00020.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00006-of-00020.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00006-of-00020.safetensors", "model.layers.90.input_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.56.input_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.22.input_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.90.mlp.down_proj.weight": "model-00006-of-00020.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00006-of-00020.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00006-of-00020.safetensors", "model.layers.90.mlp.gate_proj.weight": "model-00006-of-00020.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00006-of-00020.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00006-of-00020.safetensors", "model.layers.90.mlp.up_proj.weight": "model-00006-of-00020.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00006-of-00020.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00006-of-00020.safetensors", "model.layers.90.post_attention_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.90.self_attn.k_proj.weight": "model-00006-of-00020.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00006-of-00020.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00006-of-00020.safetensors", "model.layers.90.self_attn.o_proj.weight": "model-00006-of-00020.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00006-of-00020.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00006-of-00020.safetensors", "model.layers.90.self_attn.q_proj.weight": "model-00006-of-00020.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00006-of-00020.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00006-of-00020.safetensors", "model.layers.90.self_attn.v_proj.weight": "model-00006-of-00020.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00006-of-00020.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00006-of-00020.safetensors", "model.layers.91.input_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.57.input_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.23.input_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.91.mlp.down_proj.weight": "model-00006-of-00020.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00006-of-00020.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00006-of-00020.safetensors", "model.layers.91.mlp.gate_proj.weight": "model-00006-of-00020.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00006-of-00020.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00006-of-00020.safetensors", "model.layers.91.mlp.up_proj.weight": "model-00006-of-00020.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00006-of-00020.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00006-of-00020.safetensors", "model.layers.91.post_attention_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.91.self_attn.k_proj.weight": "model-00006-of-00020.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00006-of-00020.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00006-of-00020.safetensors", "model.layers.91.self_attn.o_proj.weight": "model-00006-of-00020.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00006-of-00020.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00006-of-00020.safetensors", "model.layers.91.self_attn.q_proj.weight": "model-00006-of-00020.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00006-of-00020.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00006-of-00020.safetensors", "model.layers.91.self_attn.v_proj.weight": "model-00006-of-00020.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00006-of-00020.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00006-of-00020.safetensors", "model.layers.92.input_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.58.input_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.24.input_layernorm.weight": "model-00006-of-00020.safetensors", "model.layers.92.mlp.down_proj.weight": "model-00007-of-00020.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00007-of-00020.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00007-of-00020.safetensors", "model.layers.92.mlp.gate_proj.weight": "model-00007-of-00020.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00007-of-00020.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00007-of-00020.safetensors", "model.layers.92.mlp.up_proj.weight": "model-00007-of-00020.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00007-of-00020.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00007-of-00020.safetensors", "model.layers.92.post_attention_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.92.self_attn.k_proj.weight": "model-00007-of-00020.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00007-of-00020.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00007-of-00020.safetensors", "model.layers.92.self_attn.o_proj.weight": "model-00007-of-00020.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00007-of-00020.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00007-of-00020.safetensors", "model.layers.92.self_attn.q_proj.weight": "model-00007-of-00020.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00007-of-00020.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00007-of-00020.safetensors", "model.layers.92.self_attn.v_proj.weight": "model-00007-of-00020.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00007-of-00020.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00007-of-00020.safetensors", "model.layers.93.input_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.59.input_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.25.input_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.93.mlp.down_proj.weight": "model-00007-of-00020.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00007-of-00020.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00007-of-00020.safetensors", "model.layers.93.mlp.gate_proj.weight": "model-00007-of-00020.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00007-of-00020.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00007-of-00020.safetensors", "model.layers.93.mlp.up_proj.weight": "model-00007-of-00020.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00007-of-00020.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00007-of-00020.safetensors", "model.layers.93.post_attention_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.93.self_attn.k_proj.weight": "model-00007-of-00020.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00007-of-00020.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00007-of-00020.safetensors", "model.layers.93.self_attn.o_proj.weight": "model-00007-of-00020.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00007-of-00020.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00007-of-00020.safetensors", "model.layers.93.self_attn.q_proj.weight": "model-00007-of-00020.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00007-of-00020.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00007-of-00020.safetensors", "model.layers.93.self_attn.v_proj.weight": "model-00007-of-00020.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00007-of-00020.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00007-of-00020.safetensors", "model.layers.94.input_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.60.input_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.26.input_layernorm.weight": "model-00007-of-00020.safetensors", "model.layers.94.mlp.down_proj.weight": "model-00007-of-00020.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00008-of-00020.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00008-of-00020.safetensors", "model.layers.94.mlp.gate_proj.weight": "model-00008-of-00020.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00008-of-00020.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00008-of-00020.safetensors", "model.layers.94.mlp.up_proj.weight": "model-00008-of-00020.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00008-of-00020.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00008-of-00020.safetensors", "model.layers.94.post_attention_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.94.self_attn.k_proj.weight": "model-00008-of-00020.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00008-of-00020.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00008-of-00020.safetensors", "model.layers.94.self_attn.o_proj.weight": "model-00008-of-00020.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00008-of-00020.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00008-of-00020.safetensors", "model.layers.94.self_attn.q_proj.weight": "model-00008-of-00020.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00008-of-00020.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00008-of-00020.safetensors", "model.layers.94.self_attn.v_proj.weight": "model-00008-of-00020.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00008-of-00020.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00008-of-00020.safetensors", "model.layers.95.input_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.61.input_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.27.input_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.95.mlp.down_proj.weight": "model-00008-of-00020.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00008-of-00020.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00008-of-00020.safetensors", "model.layers.95.mlp.gate_proj.weight": "model-00008-of-00020.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00008-of-00020.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00008-of-00020.safetensors", "model.layers.95.mlp.up_proj.weight": "model-00008-of-00020.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00008-of-00020.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00008-of-00020.safetensors", "model.layers.95.post_attention_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.95.self_attn.k_proj.weight": "model-00008-of-00020.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00008-of-00020.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00008-of-00020.safetensors", "model.layers.95.self_attn.o_proj.weight": "model-00008-of-00020.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00008-of-00020.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00008-of-00020.safetensors", "model.layers.95.self_attn.q_proj.weight": "model-00008-of-00020.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00008-of-00020.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00008-of-00020.safetensors", "model.layers.95.self_attn.v_proj.weight": "model-00008-of-00020.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00008-of-00020.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00008-of-00020.safetensors", "model.layers.96.input_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.62.input_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.28.input_layernorm.weight": "model-00008-of-00020.safetensors", "model.layers.96.mlp.down_proj.weight": "model-00008-of-00020.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00008-of-00020.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00009-of-00020.safetensors", "model.layers.96.mlp.gate_proj.weight": "model-00009-of-00020.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00009-of-00020.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00009-of-00020.safetensors", "model.layers.96.mlp.up_proj.weight": "model-00009-of-00020.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00009-of-00020.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00009-of-00020.safetensors", "model.layers.96.post_attention_layernorm.weight": "model-00009-of-00020.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00009-of-00020.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00009-of-00020.safetensors", "model.layers.96.self_attn.k_proj.weight": "model-00009-of-00020.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00009-of-00020.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00009-of-00020.safetensors", "model.layers.96.self_attn.o_proj.weight": "model-00009-of-00020.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00009-of-00020.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00009-of-00020.safetensors", "model.layers.96.self_attn.q_proj.weight": "model-00009-of-00020.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00009-of-00020.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00009-of-00020.safetensors", "model.layers.96.self_attn.v_proj.weight": "model-00009-of-00020.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00009-of-00020.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00009-of-00020.safetensors", "model.layers.97.input_layernorm.weight": "model-00009-of-00020.safetensors", "model.layers.63.input_layernorm.weight": "model-00009-of-00020.safetensors", "model.layers.29.input_layernorm.weight": "model-00009-of-00020.safetensors", "model.layers.97.mlp.down_proj.weight": "model-00009-of-00020.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00009-of-00020.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00009-of-00020.safetensors", "model.layers.97.mlp.gate_proj.weight": "model-00009-of-00020.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00009-of-00020.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00009-of-00020.safetensors", "model.layers.97.mlp.up_proj.weight": "model-00009-of-00020.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00009-of-00020.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00009-of-00020.safetensors", "model.layers.97.post_attention_layernorm.weight": "model-00009-of-00020.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00009-of-00020.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00009-of-00020.safetensors", "model.layers.97.self_attn.k_proj.weight": "model-00009-of-00020.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00009-of-00020.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00009-of-00020.safetensors", "model.layers.97.self_attn.o_proj.weight": "model-00009-of-00020.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00009-of-00020.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00009-of-00020.safetensors", "model.layers.97.self_attn.q_proj.weight": "model-00009-of-00020.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00009-of-00020.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00009-of-00020.safetensors", "model.layers.97.self_attn.v_proj.weight": "model-00009-of-00020.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00009-of-00020.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00009-of-00020.safetensors", "model.layers.3.input_layernorm.weight": "model-00009-of-00020.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00009-of-00020.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00009-of-00020.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00009-of-00020.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00009-of-00020.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00009-of-00020.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00009-of-00020.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00010-of-00020.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00010-of-00020.safetensors", "model.layers.98.input_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.64.input_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.30.input_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.98.mlp.down_proj.weight": "model-00010-of-00020.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00010-of-00020.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00010-of-00020.safetensors", "model.layers.98.mlp.gate_proj.weight": "model-00010-of-00020.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00010-of-00020.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00010-of-00020.safetensors", "model.layers.98.mlp.up_proj.weight": "model-00010-of-00020.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00010-of-00020.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00010-of-00020.safetensors", "model.layers.98.post_attention_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.98.self_attn.k_proj.weight": "model-00010-of-00020.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00010-of-00020.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00010-of-00020.safetensors", "model.layers.98.self_attn.o_proj.weight": "model-00010-of-00020.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00010-of-00020.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00010-of-00020.safetensors", "model.layers.98.self_attn.q_proj.weight": "model-00010-of-00020.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00010-of-00020.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00010-of-00020.safetensors", "model.layers.98.self_attn.v_proj.weight": "model-00010-of-00020.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00010-of-00020.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00010-of-00020.safetensors", "model.layers.99.input_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.65.input_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.31.input_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.99.mlp.down_proj.weight": "model-00010-of-00020.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00010-of-00020.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00010-of-00020.safetensors", "model.layers.99.mlp.gate_proj.weight": "model-00010-of-00020.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00010-of-00020.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00010-of-00020.safetensors", "model.layers.99.mlp.up_proj.weight": "model-00010-of-00020.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00010-of-00020.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00010-of-00020.safetensors", "model.layers.99.post_attention_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.99.self_attn.k_proj.weight": "model-00010-of-00020.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00010-of-00020.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00010-of-00020.safetensors", "model.layers.99.self_attn.o_proj.weight": "model-00010-of-00020.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00010-of-00020.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00010-of-00020.safetensors", "model.layers.99.self_attn.q_proj.weight": "model-00010-of-00020.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00010-of-00020.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00010-of-00020.safetensors", "model.layers.99.self_attn.v_proj.weight": "model-00010-of-00020.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00010-of-00020.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00010-of-00020.safetensors", "model.layers.100.input_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.66.input_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.32.input_layernorm.weight": "model-00010-of-00020.safetensors", "model.layers.100.mlp.down_proj.weight": "model-00010-of-00020.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00011-of-00020.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00011-of-00020.safetensors", "model.layers.100.mlp.gate_proj.weight": "model-00011-of-00020.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00011-of-00020.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00011-of-00020.safetensors", "model.layers.100.mlp.up_proj.weight": "model-00011-of-00020.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00011-of-00020.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00011-of-00020.safetensors", "model.layers.100.post_attention_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.100.self_attn.k_proj.weight": "model-00011-of-00020.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00011-of-00020.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00011-of-00020.safetensors", "model.layers.100.self_attn.o_proj.weight": "model-00011-of-00020.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00011-of-00020.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00011-of-00020.safetensors", "model.layers.100.self_attn.q_proj.weight": "model-00011-of-00020.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00011-of-00020.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00011-of-00020.safetensors", "model.layers.100.self_attn.v_proj.weight": "model-00011-of-00020.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00011-of-00020.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00011-of-00020.safetensors", "model.layers.101.input_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.67.input_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.33.input_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.101.mlp.down_proj.weight": "model-00011-of-00020.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00011-of-00020.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00011-of-00020.safetensors", "model.layers.101.mlp.gate_proj.weight": "model-00011-of-00020.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00011-of-00020.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00011-of-00020.safetensors", "model.layers.101.mlp.up_proj.weight": "model-00011-of-00020.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00011-of-00020.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00011-of-00020.safetensors", "model.layers.101.post_attention_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.101.self_attn.k_proj.weight": "model-00011-of-00020.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00011-of-00020.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00011-of-00020.safetensors", "model.layers.101.self_attn.o_proj.weight": "model-00011-of-00020.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00011-of-00020.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00011-of-00020.safetensors", "model.layers.101.self_attn.q_proj.weight": "model-00011-of-00020.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00011-of-00020.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00011-of-00020.safetensors", "model.layers.101.self_attn.v_proj.weight": "model-00011-of-00020.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00011-of-00020.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00011-of-00020.safetensors", "model.layers.102.input_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.68.input_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.34.input_layernorm.weight": "model-00011-of-00020.safetensors", "model.layers.102.mlp.down_proj.weight": "model-00011-of-00020.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00011-of-00020.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00012-of-00020.safetensors", "model.layers.102.mlp.gate_proj.weight": "model-00012-of-00020.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00012-of-00020.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00012-of-00020.safetensors", "model.layers.102.mlp.up_proj.weight": "model-00012-of-00020.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00012-of-00020.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00012-of-00020.safetensors", "model.layers.102.post_attention_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.102.self_attn.k_proj.weight": "model-00012-of-00020.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00012-of-00020.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00012-of-00020.safetensors", "model.layers.102.self_attn.o_proj.weight": "model-00012-of-00020.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00012-of-00020.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00012-of-00020.safetensors", "model.layers.102.self_attn.q_proj.weight": "model-00012-of-00020.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00012-of-00020.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00012-of-00020.safetensors", "model.layers.102.self_attn.v_proj.weight": "model-00012-of-00020.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00012-of-00020.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00012-of-00020.safetensors", "model.layers.103.input_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.69.input_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.35.input_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.103.mlp.down_proj.weight": "model-00012-of-00020.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00012-of-00020.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00012-of-00020.safetensors", "model.layers.103.mlp.gate_proj.weight": "model-00012-of-00020.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00012-of-00020.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00012-of-00020.safetensors", "model.layers.103.mlp.up_proj.weight": "model-00012-of-00020.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00012-of-00020.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00012-of-00020.safetensors", "model.layers.103.post_attention_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.103.self_attn.k_proj.weight": "model-00012-of-00020.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00012-of-00020.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00012-of-00020.safetensors", "model.layers.103.self_attn.o_proj.weight": "model-00012-of-00020.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00012-of-00020.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00012-of-00020.safetensors", "model.layers.103.self_attn.q_proj.weight": "model-00012-of-00020.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00012-of-00020.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00012-of-00020.safetensors", "model.layers.103.self_attn.v_proj.weight": "model-00012-of-00020.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00012-of-00020.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00012-of-00020.safetensors", "model.layers.104.input_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.70.input_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.36.input_layernorm.weight": "model-00012-of-00020.safetensors", "model.layers.104.mlp.down_proj.weight": "model-00012-of-00020.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00012-of-00020.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00012-of-00020.safetensors", "model.layers.104.mlp.gate_proj.weight": "model-00013-of-00020.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00013-of-00020.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00013-of-00020.safetensors", "model.layers.104.mlp.up_proj.weight": "model-00013-of-00020.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00013-of-00020.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00013-of-00020.safetensors", "model.layers.104.post_attention_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.104.self_attn.k_proj.weight": "model-00013-of-00020.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00013-of-00020.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00013-of-00020.safetensors", "model.layers.104.self_attn.o_proj.weight": "model-00013-of-00020.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00013-of-00020.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00013-of-00020.safetensors", "model.layers.104.self_attn.q_proj.weight": "model-00013-of-00020.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00013-of-00020.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00013-of-00020.safetensors", "model.layers.104.self_attn.v_proj.weight": "model-00013-of-00020.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00013-of-00020.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00013-of-00020.safetensors", "model.layers.105.input_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.71.input_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.37.input_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.105.mlp.down_proj.weight": "model-00013-of-00020.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00013-of-00020.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00013-of-00020.safetensors", "model.layers.105.mlp.gate_proj.weight": "model-00013-of-00020.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00013-of-00020.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00013-of-00020.safetensors", "model.layers.105.mlp.up_proj.weight": "model-00013-of-00020.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00013-of-00020.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00013-of-00020.safetensors", "model.layers.105.post_attention_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.105.self_attn.k_proj.weight": "model-00013-of-00020.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00013-of-00020.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00013-of-00020.safetensors", "model.layers.105.self_attn.o_proj.weight": "model-00013-of-00020.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00013-of-00020.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00013-of-00020.safetensors", "model.layers.105.self_attn.q_proj.weight": "model-00013-of-00020.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00013-of-00020.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00013-of-00020.safetensors", "model.layers.105.self_attn.v_proj.weight": "model-00013-of-00020.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00013-of-00020.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00013-of-00020.safetensors", "model.layers.106.input_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.72.input_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.38.input_layernorm.weight": "model-00013-of-00020.safetensors", "model.layers.106.mlp.down_proj.weight": "model-00013-of-00020.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00013-of-00020.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00013-of-00020.safetensors", "model.layers.106.mlp.gate_proj.weight": "model-00013-of-00020.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00014-of-00020.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00014-of-00020.safetensors", "model.layers.106.mlp.up_proj.weight": "model-00014-of-00020.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00014-of-00020.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00014-of-00020.safetensors", "model.layers.106.post_attention_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.106.self_attn.k_proj.weight": "model-00014-of-00020.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00014-of-00020.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00014-of-00020.safetensors", "model.layers.106.self_attn.o_proj.weight": "model-00014-of-00020.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00014-of-00020.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00014-of-00020.safetensors", "model.layers.106.self_attn.q_proj.weight": "model-00014-of-00020.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00014-of-00020.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00014-of-00020.safetensors", "model.layers.106.self_attn.v_proj.weight": "model-00014-of-00020.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00014-of-00020.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00014-of-00020.safetensors", "model.layers.107.input_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.73.input_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.39.input_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.107.mlp.down_proj.weight": "model-00014-of-00020.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00014-of-00020.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00014-of-00020.safetensors", "model.layers.107.mlp.gate_proj.weight": "model-00014-of-00020.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00014-of-00020.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00014-of-00020.safetensors", "model.layers.107.mlp.up_proj.weight": "model-00014-of-00020.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00014-of-00020.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00014-of-00020.safetensors", "model.layers.107.post_attention_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.107.self_attn.k_proj.weight": "model-00014-of-00020.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00014-of-00020.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00014-of-00020.safetensors", "model.layers.107.self_attn.o_proj.weight": "model-00014-of-00020.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00014-of-00020.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00014-of-00020.safetensors", "model.layers.107.self_attn.q_proj.weight": "model-00014-of-00020.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00014-of-00020.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00014-of-00020.safetensors", "model.layers.107.self_attn.v_proj.weight": "model-00014-of-00020.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00014-of-00020.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00014-of-00020.safetensors", "model.layers.4.input_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00014-of-00020.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00014-of-00020.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00014-of-00020.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00014-of-00020.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00014-of-00020.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00014-of-00020.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00014-of-00020.safetensors", "model.layers.108.input_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.74.input_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.40.input_layernorm.weight": "model-00014-of-00020.safetensors", "model.layers.108.mlp.down_proj.weight": "model-00014-of-00020.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00015-of-00020.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00015-of-00020.safetensors", "model.layers.108.mlp.gate_proj.weight": "model-00015-of-00020.safetensors", "model.layers.74.mlp.gate_proj.weight": "model-00015-of-00020.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00015-of-00020.safetensors", "model.layers.108.mlp.up_proj.weight": "model-00015-of-00020.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00015-of-00020.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00015-of-00020.safetensors", "model.layers.108.post_attention_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.108.self_attn.k_proj.weight": "model-00015-of-00020.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00015-of-00020.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00015-of-00020.safetensors", "model.layers.108.self_attn.o_proj.weight": "model-00015-of-00020.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00015-of-00020.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00015-of-00020.safetensors", "model.layers.108.self_attn.q_proj.weight": "model-00015-of-00020.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00015-of-00020.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00015-of-00020.safetensors", "model.layers.108.self_attn.v_proj.weight": "model-00015-of-00020.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00015-of-00020.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00015-of-00020.safetensors", "model.layers.109.input_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.75.input_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.41.input_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.109.mlp.down_proj.weight": "model-00015-of-00020.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00015-of-00020.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00015-of-00020.safetensors", "model.layers.109.mlp.gate_proj.weight": "model-00015-of-00020.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00015-of-00020.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00015-of-00020.safetensors", "model.layers.109.mlp.up_proj.weight": "model-00015-of-00020.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00015-of-00020.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00015-of-00020.safetensors", "model.layers.109.post_attention_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.109.self_attn.k_proj.weight": "model-00015-of-00020.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00015-of-00020.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00015-of-00020.safetensors", "model.layers.109.self_attn.o_proj.weight": "model-00015-of-00020.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00015-of-00020.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00015-of-00020.safetensors", "model.layers.109.self_attn.q_proj.weight": "model-00015-of-00020.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00015-of-00020.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00015-of-00020.safetensors", "model.layers.109.self_attn.v_proj.weight": "model-00015-of-00020.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00015-of-00020.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00015-of-00020.safetensors", "model.layers.110.input_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.76.input_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.42.input_layernorm.weight": "model-00015-of-00020.safetensors", "model.layers.110.mlp.down_proj.weight": "model-00015-of-00020.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00015-of-00020.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00016-of-00020.safetensors", "model.layers.110.mlp.gate_proj.weight": "model-00016-of-00020.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00016-of-00020.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00016-of-00020.safetensors", "model.layers.110.mlp.up_proj.weight": "model-00016-of-00020.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00016-of-00020.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00016-of-00020.safetensors", "model.layers.110.post_attention_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.110.self_attn.k_proj.weight": "model-00016-of-00020.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00016-of-00020.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00016-of-00020.safetensors", "model.layers.110.self_attn.o_proj.weight": "model-00016-of-00020.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00016-of-00020.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00016-of-00020.safetensors", "model.layers.110.self_attn.q_proj.weight": "model-00016-of-00020.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00016-of-00020.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00016-of-00020.safetensors", "model.layers.110.self_attn.v_proj.weight": "model-00016-of-00020.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00016-of-00020.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00016-of-00020.safetensors", "model.layers.111.input_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.77.input_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.43.input_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.111.mlp.down_proj.weight": "model-00016-of-00020.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00016-of-00020.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00016-of-00020.safetensors", "model.layers.111.mlp.gate_proj.weight": "model-00016-of-00020.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00016-of-00020.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00016-of-00020.safetensors", "model.layers.111.mlp.up_proj.weight": "model-00016-of-00020.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00016-of-00020.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00016-of-00020.safetensors", "model.layers.111.post_attention_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.111.self_attn.k_proj.weight": "model-00016-of-00020.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00016-of-00020.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00016-of-00020.safetensors", "model.layers.111.self_attn.o_proj.weight": "model-00016-of-00020.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00016-of-00020.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00016-of-00020.safetensors", "model.layers.111.self_attn.q_proj.weight": "model-00016-of-00020.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00016-of-00020.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00016-of-00020.safetensors", "model.layers.111.self_attn.v_proj.weight": "model-00016-of-00020.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00016-of-00020.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00016-of-00020.safetensors", "model.layers.112.input_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.78.input_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.44.input_layernorm.weight": "model-00016-of-00020.safetensors", "model.layers.112.mlp.down_proj.weight": "model-00016-of-00020.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00016-of-00020.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00016-of-00020.safetensors", "model.layers.112.mlp.gate_proj.weight": "model-00017-of-00020.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00017-of-00020.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00017-of-00020.safetensors", "model.layers.112.mlp.up_proj.weight": "model-00017-of-00020.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00017-of-00020.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00017-of-00020.safetensors", "model.layers.112.post_attention_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.112.self_attn.k_proj.weight": "model-00017-of-00020.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00017-of-00020.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00017-of-00020.safetensors", "model.layers.112.self_attn.o_proj.weight": "model-00017-of-00020.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00017-of-00020.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00017-of-00020.safetensors", "model.layers.112.self_attn.q_proj.weight": "model-00017-of-00020.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00017-of-00020.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00017-of-00020.safetensors", "model.layers.112.self_attn.v_proj.weight": "model-00017-of-00020.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00017-of-00020.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00017-of-00020.safetensors", "model.layers.113.input_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.79.input_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.45.input_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.113.mlp.down_proj.weight": "model-00017-of-00020.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00017-of-00020.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00017-of-00020.safetensors", "model.layers.113.mlp.gate_proj.weight": "model-00017-of-00020.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00017-of-00020.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00017-of-00020.safetensors", "model.layers.113.mlp.up_proj.weight": "model-00017-of-00020.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00017-of-00020.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00017-of-00020.safetensors", "model.layers.113.post_attention_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.113.self_attn.k_proj.weight": "model-00017-of-00020.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00017-of-00020.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00017-of-00020.safetensors", "model.layers.113.self_attn.o_proj.weight": "model-00017-of-00020.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00017-of-00020.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00017-of-00020.safetensors", "model.layers.113.self_attn.q_proj.weight": "model-00017-of-00020.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00017-of-00020.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00017-of-00020.safetensors", "model.layers.113.self_attn.v_proj.weight": "model-00017-of-00020.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00017-of-00020.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00017-of-00020.safetensors", "model.layers.114.input_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.80.input_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.46.input_layernorm.weight": "model-00017-of-00020.safetensors", "model.layers.114.mlp.down_proj.weight": "model-00017-of-00020.safetensors", "model.layers.80.mlp.down_proj.weight": "model-00017-of-00020.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00017-of-00020.safetensors", "model.layers.114.mlp.gate_proj.weight": "model-00017-of-00020.safetensors", "model.layers.80.mlp.gate_proj.weight": "model-00018-of-00020.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00018-of-00020.safetensors", "model.layers.114.mlp.up_proj.weight": "model-00018-of-00020.safetensors", "model.layers.80.mlp.up_proj.weight": "model-00018-of-00020.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00018-of-00020.safetensors", "model.layers.114.post_attention_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.80.post_attention_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.114.self_attn.k_proj.weight": "model-00018-of-00020.safetensors", "model.layers.80.self_attn.k_proj.weight": "model-00018-of-00020.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00018-of-00020.safetensors", "model.layers.114.self_attn.o_proj.weight": "model-00018-of-00020.safetensors", "model.layers.80.self_attn.o_proj.weight": "model-00018-of-00020.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00018-of-00020.safetensors", "model.layers.114.self_attn.q_proj.weight": "model-00018-of-00020.safetensors", "model.layers.80.self_attn.q_proj.weight": "model-00018-of-00020.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00018-of-00020.safetensors", "model.layers.114.self_attn.v_proj.weight": "model-00018-of-00020.safetensors", "model.layers.80.self_attn.v_proj.weight": "model-00018-of-00020.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00018-of-00020.safetensors", "model.layers.115.input_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.81.input_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.47.input_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.115.mlp.down_proj.weight": "model-00018-of-00020.safetensors", "model.layers.81.mlp.down_proj.weight": "model-00018-of-00020.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00018-of-00020.safetensors", "model.layers.115.mlp.gate_proj.weight": "model-00018-of-00020.safetensors", "model.layers.81.mlp.gate_proj.weight": "model-00018-of-00020.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00018-of-00020.safetensors", "model.layers.115.mlp.up_proj.weight": "model-00018-of-00020.safetensors", "model.layers.81.mlp.up_proj.weight": "model-00018-of-00020.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00018-of-00020.safetensors", "model.layers.115.post_attention_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.81.post_attention_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.115.self_attn.k_proj.weight": "model-00018-of-00020.safetensors", "model.layers.81.self_attn.k_proj.weight": "model-00018-of-00020.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00018-of-00020.safetensors", "model.layers.115.self_attn.o_proj.weight": "model-00018-of-00020.safetensors", "model.layers.81.self_attn.o_proj.weight": "model-00018-of-00020.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00018-of-00020.safetensors", "model.layers.115.self_attn.q_proj.weight": "model-00018-of-00020.safetensors", "model.layers.81.self_attn.q_proj.weight": "model-00018-of-00020.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00018-of-00020.safetensors", "model.layers.115.self_attn.v_proj.weight": "model-00018-of-00020.safetensors", "model.layers.81.self_attn.v_proj.weight": "model-00018-of-00020.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00018-of-00020.safetensors", "model.layers.116.input_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.116.mlp.down_proj.weight": "model-00018-of-00020.safetensors", "model.layers.116.mlp.gate_proj.weight": "model-00018-of-00020.safetensors", "model.layers.116.mlp.up_proj.weight": "model-00018-of-00020.safetensors", "model.layers.116.post_attention_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.116.self_attn.k_proj.weight": "model-00018-of-00020.safetensors", "model.layers.116.self_attn.o_proj.weight": "model-00018-of-00020.safetensors", "model.layers.116.self_attn.q_proj.weight": "model-00018-of-00020.safetensors", "model.layers.116.self_attn.v_proj.weight": "model-00018-of-00020.safetensors", "model.layers.117.input_layernorm.weight": "model-00018-of-00020.safetensors", "model.layers.117.mlp.down_proj.weight": "model-00018-of-00020.safetensors", "model.layers.117.mlp.gate_proj.weight": "model-00019-of-00020.safetensors", "model.layers.117.mlp.up_proj.weight": "model-00019-of-00020.safetensors", "model.layers.117.post_attention_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.117.self_attn.k_proj.weight": "model-00019-of-00020.safetensors", "model.layers.117.self_attn.o_proj.weight": "model-00019-of-00020.safetensors", "model.layers.117.self_attn.q_proj.weight": "model-00019-of-00020.safetensors", "model.layers.117.self_attn.v_proj.weight": "model-00019-of-00020.safetensors", "model.layers.5.input_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00019-of-00020.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00019-of-00020.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00019-of-00020.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00019-of-00020.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00019-of-00020.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00019-of-00020.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00019-of-00020.safetensors", "model.layers.118.input_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.118.mlp.down_proj.weight": "model-00019-of-00020.safetensors", "model.layers.118.mlp.gate_proj.weight": "model-00019-of-00020.safetensors", "model.layers.118.mlp.up_proj.weight": "model-00019-of-00020.safetensors", "model.layers.118.post_attention_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.118.self_attn.k_proj.weight": "model-00019-of-00020.safetensors", "model.layers.118.self_attn.o_proj.weight": "model-00019-of-00020.safetensors", "model.layers.118.self_attn.q_proj.weight": "model-00019-of-00020.safetensors", "model.layers.118.self_attn.v_proj.weight": "model-00019-of-00020.safetensors", "model.layers.119.input_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.119.mlp.down_proj.weight": "model-00019-of-00020.safetensors", "model.layers.119.mlp.gate_proj.weight": "model-00019-of-00020.safetensors", "model.layers.119.mlp.up_proj.weight": "model-00019-of-00020.safetensors", "model.layers.119.post_attention_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.119.self_attn.k_proj.weight": "model-00019-of-00020.safetensors", "model.layers.119.self_attn.o_proj.weight": "model-00019-of-00020.safetensors", "model.layers.119.self_attn.q_proj.weight": "model-00019-of-00020.safetensors", "model.layers.119.self_attn.v_proj.weight": "model-00019-of-00020.safetensors", "model.layers.120.input_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.120.mlp.down_proj.weight": "model-00019-of-00020.safetensors", "model.layers.120.mlp.gate_proj.weight": "model-00019-of-00020.safetensors", "model.layers.120.mlp.up_proj.weight": "model-00019-of-00020.safetensors", "model.layers.120.post_attention_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.120.self_attn.k_proj.weight": "model-00019-of-00020.safetensors", "model.layers.120.self_attn.o_proj.weight": "model-00019-of-00020.safetensors", "model.layers.120.self_attn.q_proj.weight": "model-00019-of-00020.safetensors", "model.layers.120.self_attn.v_proj.weight": "model-00019-of-00020.safetensors", "model.layers.121.input_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.121.mlp.down_proj.weight": "model-00019-of-00020.safetensors", "model.layers.121.mlp.gate_proj.weight": "model-00019-of-00020.safetensors", "model.layers.121.mlp.up_proj.weight": "model-00019-of-00020.safetensors", "model.layers.121.post_attention_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.121.self_attn.k_proj.weight": "model-00019-of-00020.safetensors", "model.layers.121.self_attn.o_proj.weight": "model-00019-of-00020.safetensors", "model.layers.121.self_attn.q_proj.weight": "model-00019-of-00020.safetensors", "model.layers.121.self_attn.v_proj.weight": "model-00019-of-00020.safetensors", "model.layers.122.input_layernorm.weight": "model-00019-of-00020.safetensors", "model.layers.122.mlp.down_proj.weight": "model-00019-of-00020.safetensors", "model.layers.122.mlp.gate_proj.weight": "model-00019-of-00020.safetensors", "model.layers.122.mlp.up_proj.weight": "model-00020-of-00020.safetensors", "model.layers.122.post_attention_layernorm.weight": "model-00020-of-00020.safetensors", "model.layers.122.self_attn.k_proj.weight": "model-00020-of-00020.safetensors", "model.layers.122.self_attn.o_proj.weight": "model-00020-of-00020.safetensors", "model.layers.122.self_attn.q_proj.weight": "model-00020-of-00020.safetensors", "model.layers.122.self_attn.v_proj.weight": "model-00020-of-00020.safetensors", "model.layers.6.input_layernorm.weight": "model-00020-of-00020.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00020-of-00020.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00020-of-00020.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00020-of-00020.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00020-of-00020.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00020-of-00020.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00020-of-00020.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00020-of-00020.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00020-of-00020.safetensors", "model.layers.7.input_layernorm.weight": "model-00020-of-00020.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00020-of-00020.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00020-of-00020.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00020-of-00020.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00020-of-00020.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00020-of-00020.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00020-of-00020.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00020-of-00020.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00020-of-00020.safetensors", "model.layers.8.input_layernorm.weight": "model-00020-of-00020.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00020-of-00020.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00020-of-00020.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00020-of-00020.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00020-of-00020.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00020-of-00020.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00020-of-00020.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00020-of-00020.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00020-of-00020.safetensors", "model.layers.9.input_layernorm.weight": "model-00020-of-00020.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00020-of-00020.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00020-of-00020.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00020-of-00020.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00020-of-00020.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00020-of-00020.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00020-of-00020.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00020-of-00020.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00020-of-00020.safetensors", "model.norm.weight": "model-00020-of-00020.safetensors"}}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[control_748]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59f95e28944c062244741268596badc900df86c7f5ded05088d2da22a7379e06
+size 587583