Upload folder using huggingface_hub

README.md

@@ -0,0 +1,29 @@
+---
+license: apache-2.0
+tags:
+- merge
+- mergekit
+- lazymergekit
+- WizardLM/WizardCoder-33B-V1.1
+- Phind/Phind-CodeLlama-34B-v2
+---
+
+# wizardphind-coder-passthrough-46B
+
+wizardphind-coder-passthrough-46B is a merge of the following models using [mergekit](https://github.com/cg123/mergekit):
+* [WizardLM/WizardCoder-33B-V1.1](https://huggingface.co/WizardLM/WizardCoder-33B-V1.1)
+* [Phind/Phind-CodeLlama-34B-v2](https://huggingface.co/Phind/Phind-CodeLlama-34B-v2)
+
+## 🧩 Configuration
+
+\```yaml
+slices:
+  - sources:
+    - model: WizardLM/WizardCoder-33B-V1.1
+      layer_range: [0, 62]
+  - sources:
+    - model: Phind/Phind-CodeLlama-34B-v2
+      layer_range: [24, 32]
+merge_method: passthrough
+dtype: bfloat16
+\```

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "Phind/Phind-CodeLlama-34B-v2",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 22016,
+  "max_position_embeddings": 16384,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 70,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 1000000,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.35.2",
+  "use_cache": true,
+  "vocab_size": 32000
+}

mergekit_config.yml

@@ -0,0 +1,10 @@
+
+slices:
+  - sources:
+    - model: WizardLM/WizardCoder-33B-V1.1
+      layer_range: [0, 62]
+  - sources:
+    - model: Phind/Phind-CodeLlama-34B-v2
+      layer_range: [24, 32]
+merge_method: passthrough
+dtype: bfloat16

model-00001-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8d1daf8e292a0a1b4f5c7a78e9995bc86df218c08fd76e7ea843c66ebf63b98
+size 1798352248

model-00002-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0ef271a0bcbc363ba25b6685a5ad266301c9636781304504c93d5bb56f5f3fa
+size 1846077376

model-00003-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b532cba93a0b8f2e884e7d9c7cb2395730893e3ccf4a36d4d0d154117e15e6c3
+size 1846048480

model-00004-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f95897b9089e55704a6d09d4fe795afca25df69c073d5e4f27a762c2dcee5b9
+size 1886447208

model-00005-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1bf337da63e983c48a81af7c7d4af1cc11d21494bafea6a80378902c1966b755
+size 1846062936

model-00006-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:145875e41c50ca9064139846ef086ccdf0d5183b3d80f88ab967170004ab87b4
+size 1846062928

model-00007-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3841e2d81941767fbd3e30a67e5d8d43a3a0182707339ef8218e386737958712
+size 1963503696

model-00008-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a0e2fc07b91cdb012e46aa829e5ceab0789b8fca20bbec10d90c036791362e9
+size 1814112080

model-00009-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a41d0ea3232887383500378c290df209a73a5146f940df0adbfb0e329a1708bf
+size 1744864424

model-00010-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06243c4ecd109dd06fa573823d84c7ced30aef93bb21aa94b7cd28fc0c2c0ad2
+size 1895876104

model-00011-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef5dec93d40ee0cf897c043e95715bc8ec92d4b027be5a0b771d1d5f98cf4f2c
+size 1724923152

model-00012-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff725d39f1e99089d404947235b38465fc7634bf536fa85e0585f766a57650b7
+size 1886447224

model-00013-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14bf701196e11b60139895d77d38c2cfa4803fcfe2948b16dcdb68fdc3f70f91
+size 1846062952

model-00014-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca1a0c3cc27fb6ee03dffa01b9c12c42055fb70dc7125053c53cb1bbc2eb1966
+size 1969817200

model-00015-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9358a3f63e51a916ae59a88ac1685fdbe651070aeec98c930d438f1c7622eea
+size 1669383688

model-00016-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35b8f489009f259c7e49b02b886ba861235a81f748bcba3a5369809c59a39011
+size 1744864432

model-00017-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b57232a6830e6d11d1d5db0e0cf5f24eda4256a9bf89e5551393c8a18fe8cf30
+size 1849224696

model-00018-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8887ea4552d34ea9273b4b23ed996bfa397e649147eb4ab4bf08eea67d751b6
+size 1846077400

model-00019-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d352166b9210e17ff1ff3743c5b9dbb65e1cf5b9a00b499379378e24e0377c0
+size 1963503696

model-00020-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1a7eac88af7c32488625c4de8e13568bab2339e6e4910a20297e02c8776d5f4
+size 1989207776

model-00021-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec09b1af28d55cd715e6bd77b5f30bdf4dd541a38c50f15acbeac8a33cb8133f
+size 1915807576

model-00022-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30fe322bac749de04a9c8b0c012a31d271e2c9ec17ba4be893ce84cd47d12cc7
+size 1805184704

model-00023-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4d17f7dcd9effeb3ca4f00445608e721be07e29c28b2b9bbd7658d36010427b
+size 1901112952

model-00024-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40ab312afdf93d55045414478188ae475288f250f9aa70748b4326ac675b743b
+size 1831382760

model-00025-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4b925eee5f8ccbf50fe205ec409809e63b4aa69a71d4068817380d50c43e2c8
+size 1846077400

model-00026-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15879499d9020f74f2883b0bc8b9bc151257d78a160e1253fe90b346f4a2d73c
+size 1846077400

model-00027-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83155012f5a0836fc703aa0626e5d2d9b564a1fff7a5d736c409d8c80897aef9
+size 1846048496

model-00028-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71dbedba755d7e389f5e8b40e186097908054b5a14033d848c0b328a788cde13
+size 1886447224

model-00029-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7919aa0a72737a9f73f4d4b0e6b5bf1a89923bf2d6ec4fc22fe054285f153c3d
+size 1846077400

model-00030-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2e94d7956ccc8480a84561ee568b6aeab6318ebe3ead5ac7f35fae54e76b54b
+size 1846077400

model-00031-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4181dfdc287067e43e5404f9e964db7a0420d0781020a2ac8b772f6a92371827
+size 1901112952

model-00032-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fe9b06e9f8385d1b60db2a4f517c1963ab4e0e4aee2be0ffc89193b478f0b73
+size 1948809064

model-00033-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb53db3c9571d45d84354e10eb7621d93c4dd43dbf72ab8294c2cf84c6ea2e5b
+size 1728651088

model-00034-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48f92ca7e4d4d09cba4b8d469399be3234abba1881c7df7b3a0b3bf273cdbad7
+size 1846077400

model-00035-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6724408b7ef78efc0c183f847674384a6183ee3bb3dffb465160a87b252f84c
+size 1846048496

model-00036-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96338c4bedae93870bd9756536a68b37e00ce873bdcf6ff665991ff827c661bd
+size 1886447224

model-00037-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4aa6f1c6d7a55b64d74fcd20a0b6e3b3bb1d913e8c1db21524d90d10eb14f567
+size 1846062952

model-00038-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:217746df6bac1fc77561d11b1241794ffaacd69e528082557bc5fa206c88e757
+size 1846091832

model-00039-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ac6a0d9ce6fdd34b801d416d963460cc5937c8d831e7a100fb0f422c53e57cf
+size 1901112952

model-00040-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b5965cf41fec40454118c2d9fd75bfbb36942482bac93fdf87564ee0410a364
+size 1791012944

model-00041-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be35ae8bf57750316fc646a3b6db7710c2622fca150dc5cfc82703d398cd1a75
+size 1886447224

model-00042-of-00042.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b297753ec095f02d08ea1080567bccf4c2651c4e3abf4f1fb3fb4268dac226b
+size 1846062952

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.3.2"}, "weight_map": {"model.embed_tokens.weight": "model-00001-of-00042.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00042.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00001-of-00042.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00042.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00001-of-00042.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00042.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00042.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00042.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00042.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00042.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00042.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00001-of-00042.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00042.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00002-of-00042.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00042.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00042.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00002-of-00042.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00042.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00042.safetensors", "model.layers.2.input_layernorm.weight": "model-00002-of-00042.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00002-of-00042.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00002-of-00042.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00002-of-00042.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00042.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00002-of-00042.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00042.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00002-of-00042.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00002-of-00042.safetensors", "model.layers.3.input_layernorm.weight": "model-00002-of-00042.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00003-of-00042.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00003-of-00042.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00003-of-00042.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00003-of-00042.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00003-of-00042.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00003-of-00042.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00003-of-00042.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00003-of-00042.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00003-of-00042.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00003-of-00042.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00003-of-00042.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00003-of-00042.safetensors", "model.layers.4.input_layernorm.weight": "model-00003-of-00042.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00003-of-00042.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00003-of-00042.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00004-of-00042.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00004-of-00042.safetensors", "model.layers.5.input_layernorm.weight": "model-00004-of-00042.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00004-of-00042.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00004-of-00042.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00004-of-00042.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00004-of-00042.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00004-of-00042.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00004-of-00042.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00004-of-00042.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00004-of-00042.safetensors", "model.layers.6.input_layernorm.weight": "model-00004-of-00042.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00004-of-00042.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00004-of-00042.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00005-of-00042.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00005-of-00042.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00005-of-00042.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00005-of-00042.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00005-of-00042.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00005-of-00042.safetensors", "model.layers.7.input_layernorm.weight": "model-00005-of-00042.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00005-of-00042.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00005-of-00042.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00005-of-00042.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00005-of-00042.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00005-of-00042.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00005-of-00042.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00005-of-00042.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00005-of-00042.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00005-of-00042.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00006-of-00042.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00006-of-00042.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00006-of-00042.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00006-of-00042.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00006-of-00042.safetensors", "model.layers.10.input_layernorm.weight": "model-00006-of-00042.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00006-of-00042.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00006-of-00042.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00006-of-00042.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00006-of-00042.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00006-of-00042.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00006-of-00042.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00006-of-00042.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00006-of-00042.safetensors", "model.layers.11.input_layernorm.weight": "model-00006-of-00042.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00006-of-00042.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00007-of-00042.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00007-of-00042.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00007-of-00042.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00007-of-00042.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00007-of-00042.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00007-of-00042.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00007-of-00042.safetensors", "model.layers.12.input_layernorm.weight": "model-00007-of-00042.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00007-of-00042.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00007-of-00042.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00007-of-00042.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00007-of-00042.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00007-of-00042.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00007-of-00042.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00007-of-00042.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00007-of-00042.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00007-of-00042.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00007-of-00042.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00008-of-00042.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00008-of-00042.safetensors", "model.layers.8.input_layernorm.weight": "model-00008-of-00042.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00008-of-00042.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00008-of-00042.safetensors", "model.layers.9.input_layernorm.weight": "model-00008-of-00042.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00008-of-00042.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00008-of-00042.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00008-of-00042.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00008-of-00042.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00008-of-00042.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00008-of-00042.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00008-of-00042.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00008-of-00042.safetensors", "model.layers.62.input_layernorm.weight": "model-00008-of-00042.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00008-of-00042.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00009-of-00042.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00009-of-00042.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00009-of-00042.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00009-of-00042.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00009-of-00042.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00009-of-00042.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00009-of-00042.safetensors", "model.layers.63.input_layernorm.weight": "model-00009-of-00042.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00009-of-00042.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00009-of-00042.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00010-of-00042.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00010-of-00042.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00010-of-00042.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00010-of-00042.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00010-of-00042.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00010-of-00042.safetensors", "model.layers.64.input_layernorm.weight": "model-00010-of-00042.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00010-of-00042.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00010-of-00042.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00010-of-00042.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00010-of-00042.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00010-of-00042.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00010-of-00042.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00011-of-00042.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00011-of-00042.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00011-of-00042.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00011-of-00042.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00011-of-00042.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00011-of-00042.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00011-of-00042.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00011-of-00042.safetensors", "model.layers.13.input_layernorm.weight": "model-00011-of-00042.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00011-of-00042.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00011-of-00042.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00012-of-00042.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00012-of-00042.safetensors", "model.layers.14.input_layernorm.weight": "model-00012-of-00042.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00012-of-00042.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00012-of-00042.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00012-of-00042.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00012-of-00042.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00012-of-00042.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00012-of-00042.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00012-of-00042.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00012-of-00042.safetensors", "model.layers.15.input_layernorm.weight": "model-00012-of-00042.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00012-of-00042.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00012-of-00042.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00013-of-00042.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00013-of-00042.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00013-of-00042.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00013-of-00042.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00013-of-00042.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00013-of-00042.safetensors", "model.layers.16.input_layernorm.weight": "model-00013-of-00042.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00013-of-00042.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00013-of-00042.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00013-of-00042.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00013-of-00042.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00013-of-00042.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00013-of-00042.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00013-of-00042.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00013-of-00042.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00013-of-00042.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00014-of-00042.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00014-of-00042.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00014-of-00042.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00014-of-00042.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00014-of-00042.safetensors", "model.layers.65.input_layernorm.weight": "model-00014-of-00042.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00014-of-00042.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00014-of-00042.safetensors", "model.layers.66.input_layernorm.weight": "model-00014-of-00042.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00014-of-00042.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00014-of-00042.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00014-of-00042.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00014-of-00042.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00014-of-00042.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00015-of-00042.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00015-of-00042.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00015-of-00042.safetensors", "model.layers.67.input_layernorm.weight": "model-00015-of-00042.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00015-of-00042.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00015-of-00042.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00015-of-00042.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00015-of-00042.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00015-of-00042.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00015-of-00042.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00015-of-00042.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00015-of-00042.safetensors", "model.layers.68.input_layernorm.weight": "model-00015-of-00042.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00016-of-00042.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00016-of-00042.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00016-of-00042.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00016-of-00042.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00016-of-00042.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00016-of-00042.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00016-of-00042.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00016-of-00042.safetensors", "model.layers.69.input_layernorm.weight": "model-00016-of-00042.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00016-of-00042.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00017-of-00042.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00017-of-00042.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00017-of-00042.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00017-of-00042.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00017-of-00042.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00017-of-00042.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00017-of-00042.safetensors", "model.layers.17.input_layernorm.weight": "model-00017-of-00042.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00017-of-00042.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00017-of-00042.safetensors", "model.layers.18.input_layernorm.weight": "model-00017-of-00042.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00017-of-00042.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00017-of-00042.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00018-of-00042.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00018-of-00042.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00018-of-00042.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00018-of-00042.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00018-of-00042.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00018-of-00042.safetensors", "model.layers.19.input_layernorm.weight": "model-00018-of-00042.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00018-of-00042.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00018-of-00042.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00018-of-00042.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00018-of-00042.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00018-of-00042.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00018-of-00042.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00018-of-00042.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00018-of-00042.safetensors", "model.layers.20.input_layernorm.weight": "model-00018-of-00042.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00018-of-00042.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00019-of-00042.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00019-of-00042.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00019-of-00042.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00019-of-00042.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00019-of-00042.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00019-of-00042.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00019-of-00042.safetensors", "model.layers.21.input_layernorm.weight": "model-00019-of-00042.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00019-of-00042.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00019-of-00042.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00019-of-00042.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00019-of-00042.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00019-of-00042.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00019-of-00042.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00019-of-00042.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00019-of-00042.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00019-of-00042.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00019-of-00042.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00020-of-00042.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00020-of-00042.safetensors", "model.layers.22.input_layernorm.weight": "model-00020-of-00042.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00020-of-00042.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00020-of-00042.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00020-of-00042.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00020-of-00042.safetensors", "model.layers.23.input_layernorm.weight": "model-00020-of-00042.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00020-of-00042.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00020-of-00042.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00020-of-00042.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00020-of-00042.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00020-of-00042.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00020-of-00042.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00020-of-00042.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00021-of-00042.safetensors", "model.layers.24.input_layernorm.weight": "model-00021-of-00042.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00021-of-00042.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00021-of-00042.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00021-of-00042.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00021-of-00042.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00021-of-00042.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00021-of-00042.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00021-of-00042.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00021-of-00042.safetensors", "model.layers.25.input_layernorm.weight": "model-00021-of-00042.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00021-of-00042.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00021-of-00042.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00021-of-00042.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00021-of-00042.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00021-of-00042.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00022-of-00042.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00022-of-00042.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00022-of-00042.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00022-of-00042.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00022-of-00042.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00022-of-00042.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00022-of-00042.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00022-of-00042.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00022-of-00042.safetensors", "lm_head.weight": "model-00022-of-00042.safetensors", "model.norm.weight": "model-00022-of-00042.safetensors", "model.layers.26.input_layernorm.weight": "model-00022-of-00042.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00022-of-00042.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00022-of-00042.safetensors", "model.layers.27.input_layernorm.weight": "model-00022-of-00042.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00023-of-00042.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00023-of-00042.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00023-of-00042.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00023-of-00042.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00023-of-00042.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00023-of-00042.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00023-of-00042.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00023-of-00042.safetensors", "model.layers.28.input_layernorm.weight": "model-00023-of-00042.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00023-of-00042.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00023-of-00042.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00023-of-00042.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00023-of-00042.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00023-of-00042.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00024-of-00042.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00024-of-00042.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00024-of-00042.safetensors", "model.layers.29.input_layernorm.weight": "model-00024-of-00042.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00024-of-00042.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00024-of-00042.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00024-of-00042.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00024-of-00042.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00024-of-00042.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00024-of-00042.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00024-of-00042.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00024-of-00042.safetensors", "model.layers.30.input_layernorm.weight": "model-00024-of-00042.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00024-of-00042.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00024-of-00042.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00025-of-00042.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00025-of-00042.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00025-of-00042.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00025-of-00042.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00025-of-00042.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00025-of-00042.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00025-of-00042.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00025-of-00042.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00025-of-00042.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00025-of-00042.safetensors", "model.layers.31.input_layernorm.weight": "model-00025-of-00042.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00025-of-00042.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00025-of-00042.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00025-of-00042.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00025-of-00042.safetensors", "model.layers.32.input_layernorm.weight": "model-00025-of-00042.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00025-of-00042.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00026-of-00042.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00026-of-00042.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00026-of-00042.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00026-of-00042.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00026-of-00042.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00026-of-00042.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00026-of-00042.safetensors", "model.layers.33.input_layernorm.weight": "model-00026-of-00042.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00026-of-00042.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00026-of-00042.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00026-of-00042.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00026-of-00042.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00026-of-00042.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00026-of-00042.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00026-of-00042.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00026-of-00042.safetensors", "model.layers.34.input_layernorm.weight": "model-00026-of-00042.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00027-of-00042.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00027-of-00042.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00027-of-00042.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00027-of-00042.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00027-of-00042.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00027-of-00042.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00027-of-00042.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00027-of-00042.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00027-of-00042.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00027-of-00042.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00027-of-00042.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00027-of-00042.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00027-of-00042.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00027-of-00042.safetensors", "model.layers.35.input_layernorm.weight": "model-00027-of-00042.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00028-of-00042.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00028-of-00042.safetensors", "model.layers.36.input_layernorm.weight": "model-00028-of-00042.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00028-of-00042.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00028-of-00042.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00028-of-00042.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00028-of-00042.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00028-of-00042.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00028-of-00042.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00028-of-00042.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00028-of-00042.safetensors", "model.layers.37.input_layernorm.weight": "model-00028-of-00042.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00028-of-00042.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00028-of-00042.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00029-of-00042.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00029-of-00042.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00029-of-00042.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00029-of-00042.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00029-of-00042.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00029-of-00042.safetensors", "model.layers.38.input_layernorm.weight": "model-00029-of-00042.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00029-of-00042.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00029-of-00042.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00029-of-00042.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00029-of-00042.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00029-of-00042.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00029-of-00042.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00029-of-00042.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00029-of-00042.safetensors", "model.layers.39.input_layernorm.weight": "model-00029-of-00042.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00029-of-00042.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00030-of-00042.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00030-of-00042.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00030-of-00042.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00030-of-00042.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00030-of-00042.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00030-of-00042.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00030-of-00042.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00030-of-00042.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00030-of-00042.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00030-of-00042.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00030-of-00042.safetensors", "model.layers.40.input_layernorm.weight": "model-00030-of-00042.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00030-of-00042.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00030-of-00042.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00030-of-00042.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00030-of-00042.safetensors", "model.layers.41.input_layernorm.weight": "model-00030-of-00042.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00031-of-00042.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00031-of-00042.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00031-of-00042.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00031-of-00042.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00031-of-00042.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00031-of-00042.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00031-of-00042.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00031-of-00042.safetensors", "model.layers.42.input_layernorm.weight": "model-00031-of-00042.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00031-of-00042.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00031-of-00042.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00031-of-00042.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00031-of-00042.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00031-of-00042.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00032-of-00042.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00032-of-00042.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00032-of-00042.safetensors", "model.layers.43.input_layernorm.weight": "model-00032-of-00042.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00032-of-00042.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00032-of-00042.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00032-of-00042.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00032-of-00042.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00032-of-00042.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00032-of-00042.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00032-of-00042.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00032-of-00042.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00032-of-00042.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00032-of-00042.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00032-of-00042.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00032-of-00042.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00033-of-00042.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00033-of-00042.safetensors", "model.layers.44.input_layernorm.weight": "model-00033-of-00042.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00033-of-00042.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00033-of-00042.safetensors", "model.layers.45.input_layernorm.weight": "model-00033-of-00042.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00033-of-00042.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00033-of-00042.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00033-of-00042.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00033-of-00042.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00033-of-00042.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00033-of-00042.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00033-of-00042.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00033-of-00042.safetensors", "model.layers.46.input_layernorm.weight": "model-00033-of-00042.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00033-of-00042.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00034-of-00042.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00034-of-00042.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00034-of-00042.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00034-of-00042.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00034-of-00042.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00034-of-00042.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00034-of-00042.safetensors", "model.layers.47.input_layernorm.weight": "model-00034-of-00042.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00034-of-00042.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00034-of-00042.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00034-of-00042.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00034-of-00042.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00034-of-00042.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00034-of-00042.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00034-of-00042.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00034-of-00042.safetensors", "model.layers.48.input_layernorm.weight": "model-00034-of-00042.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00035-of-00042.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00035-of-00042.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00035-of-00042.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00035-of-00042.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00035-of-00042.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00035-of-00042.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00035-of-00042.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00035-of-00042.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00035-of-00042.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00035-of-00042.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00035-of-00042.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00035-of-00042.safetensors", "model.layers.49.input_layernorm.weight": "model-00035-of-00042.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00035-of-00042.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00035-of-00042.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00036-of-00042.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00036-of-00042.safetensors", "model.layers.50.input_layernorm.weight": "model-00036-of-00042.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00036-of-00042.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00036-of-00042.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00036-of-00042.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00036-of-00042.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00036-of-00042.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00036-of-00042.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00036-of-00042.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00036-of-00042.safetensors", "model.layers.51.input_layernorm.weight": "model-00036-of-00042.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00036-of-00042.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00036-of-00042.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00037-of-00042.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00037-of-00042.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00037-of-00042.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00037-of-00042.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00037-of-00042.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00037-of-00042.safetensors", "model.layers.52.input_layernorm.weight": "model-00037-of-00042.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00037-of-00042.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00037-of-00042.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00037-of-00042.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00037-of-00042.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00037-of-00042.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00037-of-00042.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00037-of-00042.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00037-of-00042.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00037-of-00042.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00038-of-00042.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00038-of-00042.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00038-of-00042.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00038-of-00042.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00038-of-00042.safetensors", "model.layers.53.input_layernorm.weight": "model-00038-of-00042.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00038-of-00042.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00038-of-00042.safetensors", "model.layers.54.input_layernorm.weight": "model-00038-of-00042.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00038-of-00042.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00038-of-00042.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00038-of-00042.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00038-of-00042.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00038-of-00042.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00038-of-00042.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00038-of-00042.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00038-of-00042.safetensors", "model.layers.55.input_layernorm.weight": "model-00038-of-00042.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00039-of-00042.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00039-of-00042.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00039-of-00042.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00039-of-00042.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00039-of-00042.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00039-of-00042.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00039-of-00042.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00039-of-00042.safetensors", "model.layers.56.input_layernorm.weight": "model-00039-of-00042.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00039-of-00042.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00039-of-00042.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00039-of-00042.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00039-of-00042.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00039-of-00042.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00040-of-00042.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00040-of-00042.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00040-of-00042.safetensors", "model.layers.57.input_layernorm.weight": "model-00040-of-00042.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00040-of-00042.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00040-of-00042.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00040-of-00042.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00040-of-00042.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00040-of-00042.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00040-of-00042.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00040-of-00042.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00040-of-00042.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00040-of-00042.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00040-of-00042.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00040-of-00042.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00040-of-00042.safetensors", "model.layers.58.input_layernorm.weight": "model-00040-of-00042.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00040-of-00042.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00041-of-00042.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00041-of-00042.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00041-of-00042.safetensors", "model.layers.59.input_layernorm.weight": "model-00041-of-00042.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00041-of-00042.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00041-of-00042.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00041-of-00042.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00041-of-00042.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00041-of-00042.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00041-of-00042.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00041-of-00042.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00041-of-00042.safetensors", "model.layers.60.input_layernorm.weight": "model-00041-of-00042.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00041-of-00042.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00042-of-00042.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00042-of-00042.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00042-of-00042.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00042-of-00042.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00042-of-00042.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00042-of-00042.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00042-of-00042.safetensors", "model.layers.61.input_layernorm.weight": "model-00042-of-00042.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00042-of-00042.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00042-of-00042.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00042-of-00042.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00042-of-00042.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00042-of-00042.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00042-of-00042.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00042-of-00042.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00042-of-00042.safetensors"}}

special_tokens_map.json

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

tokenizer.model

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

tokenizer_config.json

@@ -0,0 +1,42 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 4096,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true
+}