Upload Primary Model Files

README.md

@@ -1,3 +1,45 @@
 ---
-license: apache-2.0
+base_model:
+- Weyaxi/Einstein-v4-7B
+- NousResearch/Genstruct-7B
+- NousResearch/Nous-Hermes-2-Mistral-7B-DPO
+library_name: transformers
+tags:
+- mergekit
+- merge
+
 ---
+# mergedmodel
+
+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:
+* [Weyaxi/Einstein-v4-7B](https://huggingface.co/Weyaxi/Einstein-v4-7B)
+* [NousResearch/Genstruct-7B](https://huggingface.co/NousResearch/Genstruct-7B)
+* [NousResearch/Nous-Hermes-2-Mistral-7B-DPO](https://huggingface.co/NousResearch/Nous-Hermes-2-Mistral-7B-DPO)
+
+### Configuration
+
+The following YAML configuration was used to produce this model:
+
+```yaml
+slices:
+  - sources:
+      - model: NousResearch/Nous-Hermes-2-Mistral-7B-DPO
+        layer_range: [0, 32]
+  - sources:
+      - model: NousResearch/Genstruct-7B
+        layer_range: [0, 32]
+  - sources:
+      - model: Weyaxi/Einstein-v4-7B
+        layer_range: [0, 32]
+merge_method: passthrough
+dtype: bfloat16
+```

added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "<|im_end|>": 32000,
+  "<|im_start|>": 32001
+}

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "Weyaxi/Einstein-v4-7B",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 96,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.38.2",
+  "use_cache": false,
+  "vocab_size": 32002
+}

mergekit_config.yml

@@ -0,0 +1,12 @@
+slices:
+  - sources:
+      - model: NousResearch/Nous-Hermes-2-Mistral-7B-DPO
+        layer_range: [0, 32]
+  - sources:
+      - model: NousResearch/Genstruct-7B
+        layer_range: [0, 32]
+  - sources:
+      - model: Weyaxi/Einstein-v4-7B
+        layer_range: [0, 32]
+merge_method: passthrough
+dtype: bfloat16

model-00001-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c7c2e54bd3f70ea7f74f0d14a132a8f364e5f28968442247862e4fa5aa1b70a
+size 1979773160

model-00002-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0d441a8ce3b5f18d6d1bc675500535a352a22baf6d5958357f91ab26343215b
+size 1946235680

model-00003-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:961949d7f2d507abb9acc3e53b8b5098863a1698eb3d121ce77ded0531bab0a4
+size 1946227368

model-00004-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a88718a3a1331d1115d8b245c31ec03f149e2c8947dcf6e923b9067234f07d0
+size 1979781464

model-00005-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3962426eea84db38f27063f68a9a5b7d17efbcdac65999617427cba2cf190002
+size 1946227376

model-00006-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3933b7977d223a71339bd2f2a1baf40109bacd51d990aed8fac0225e8f9f5b71
+size 1979789736

model-00007-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4218611c922b054c9520c9e02cb4c060e0c614b25d08c02d4f4c9c10061cf2bc
+size 1946235640

model-00008-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eacf265527071a59339ba9005c8be3d6738d4dc63be6975eddc2c1f8ead9bb7
+size 1973506608

model-00009-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de2733b2698bf13b0f7e68181a2dd9f7a822eff62d77e6a7f40b90100762a275
+size 1979781464

model-00010-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:003a4c2f9fdea2813521f5aca9bb32a6214fb9ceb242aa927e5daa46cde34f32
+size 1946243992

model-00011-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78623526716dbd5dc4800e367c749b0758e8c1ce0cac6d62031366d40ce2d4ea
+size 1946243984

model-00012-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62e062d54f0ffedd3ace4a590254900d0f68f7c9945e8e18f1872ffe9b82992a
+size 1979781464

model-00013-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f304bdaed08f4364c2bd6b194cc617d9d79bdbab4ac2ddf4814c6b9c3b0415c
+size 1946244008

model-00014-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9fe7b362f95c030f9e0b352f231c3d14c1ba9536f2b7e6f4d8c243ce5c89e81
+size 1946243984

model-00015-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f13ce810a506a7c68deeeb7a7920da454d7e0d516d79f2674bea90e598be8b4e
+size 1862357432

model-00016-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:199f4f47d4d253f31b0be3581f1fcc1bca48dcaa2aab3b165ebfd493fd98f422
+size 1973506120

model-00017-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:265c9e9f59f52bf5a3d57c4456f1f263a158fc147155a982d52fab43f33f0ace
+size 1895903680

model-00018-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87936d1b135326e34ca4d7b0daac8e26f3347631e63ec89dbda21eba7e7db694
+size 1979798056

model-00019-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0de683c96676aadfd7d23b2818ecc7c1dc633666fcab6897fd54efe9fb734e49
+size 1979781464

model-00020-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:886d64ef68c9a1612498f86cc5152e204afde6eacb4fa699bf08b185393796e9
+size 1979798072

model-00021-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb69b758b098e078e353f223264b1a27f00102202db1ffbd85660ee65d7b9616
+size 1979789776

model-00022-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4042a96b323dfacd2395b613e1708f6d797afdda231b2e36a667b12ef38ec5f1
+size 1308683448

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.4.1", "total_size": 42401832960}, "weight_map": {"model.layers.74.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.73.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.72.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.71.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.70.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.69.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.68.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.67.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.66.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.65.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.64.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.41.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.40.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.39.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.38.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.37.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.36.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.35.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.34.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.33.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.32.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.9.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.8.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.7.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.6.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.5.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.4.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.3.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.2.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.1.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.0.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.embed_tokens.weight": "model-00008-of-00022.safetensors", "model.layers.86.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.86.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.86.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.86.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.85.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.85.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.85.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.85.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.85.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.85.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.85.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.85.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.85.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.84.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.84.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.84.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.84.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.84.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.84.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.84.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.84.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.84.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.83.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.83.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.83.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.83.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.83.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.83.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.83.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.83.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.83.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.82.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.82.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.82.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.82.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.82.self_attn.o_proj.weight": "model-00009-of-00022.safetensors", "model.layers.82.self_attn.v_proj.weight": "model-00009-of-00022.safetensors", "model.layers.82.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.82.self_attn.q_proj.weight": "model-00009-of-00022.safetensors", "model.layers.82.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.81.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.81.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.81.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.81.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.81.self_attn.o_proj.weight": "model-00009-of-00022.safetensors", "model.layers.81.self_attn.v_proj.weight": "model-00009-of-00022.safetensors", "model.layers.81.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.81.self_attn.q_proj.weight": "model-00009-of-00022.safetensors", "model.layers.81.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.80.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.80.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.80.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.80.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.80.self_attn.o_proj.weight": "model-00009-of-00022.safetensors", "model.layers.80.self_attn.v_proj.weight": "model-00009-of-00022.safetensors", "model.layers.80.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.80.self_attn.q_proj.weight": "model-00009-of-00022.safetensors", "model.layers.80.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00009-of-00022.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00009-of-00022.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00009-of-00022.safetensors", "model.layers.79.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.78.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.77.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.76.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.75.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.74.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.52.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.51.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.50.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.49.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.48.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.47.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.46.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.45.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.44.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.43.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.42.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.21.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.20.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.19.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.18.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.17.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.16.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.15.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.14.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.13.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.12.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.11.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.10.input_layernorm.weight": "model-00015-of-00022.safetensors", "lm_head.weight": "model-00016-of-00022.safetensors", "model.norm.weight": "model-00016-of-00022.safetensors", "model.layers.95.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.95.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.95.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.95.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.95.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.95.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.95.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.95.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.95.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.94.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.94.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.93.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.93.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.92.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.92.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.92.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.92.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.92.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.92.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.92.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.92.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.92.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.91.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.91.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.90.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.90.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.90.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.90.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.90.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.90.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.90.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.90.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.90.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.89.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.89.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.89.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.89.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.89.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.89.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.89.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.89.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.89.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.88.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.88.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.88.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.88.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.88.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.88.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.88.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.88.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.88.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.87.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.87.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.87.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.87.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.87.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.87.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.87.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.87.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.87.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.86.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.86.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.86.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.86.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.86.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.63.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.62.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.61.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.60.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.59.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.58.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.57.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.56.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.55.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.54.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.31.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.30.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.29.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.28.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.27.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.26.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00022-of-00022.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00022-of-00022.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00022-of-00022.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00022-of-00022.safetensors", "model.layers.25.input_layernorm.weight": "model-00022-of-00022.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00022-of-00022.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00022-of-00022.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00022-of-00022.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00022-of-00022.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00022-of-00022.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00022-of-00022.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00022-of-00022.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00022-of-00022.safetensors", "model.layers.24.input_layernorm.weight": "model-00022-of-00022.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00022-of-00022.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00022-of-00022.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00022-of-00022.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00022-of-00022.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00022-of-00022.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00022-of-00022.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00022-of-00022.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00022-of-00022.safetensors", "model.layers.23.input_layernorm.weight": "model-00022-of-00022.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00022-of-00022.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00022-of-00022.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00022-of-00022.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00022-of-00022.safetensors", "model.layers.22.input_layernorm.weight": "model-00022-of-00022.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": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443

tokenizer_config.json

@@ -0,0 +1,62 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32000": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = 'You are a helpful assistant.' %}{% endif %}{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in loop_messages %}{% if loop.index0 == 0 %}{{'<|im_start|>system\n' + system_message + '<|im_end|>\n'}}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "trust_remote_code": false,
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false,
+  "use_fast": true
+}