Upload folder using huggingface_hub

README.md

@@ -0,0 +1,26 @@
+---
+language:
+- en
+license: llama2
+tags:
+- mlx
+datasets:
+- ehartford/dolphin
+- jondurbin/airoboros-2.2.1
+---
+
+# mlx-community/dolphin-2.2-70b
+This model was converted to MLX format from [`cognitivecomputations/dolphin-2.2-70b`]() using mlx-lm version **0.7.0**.
+Refer to the [original model card](https://huggingface.co/cognitivecomputations/dolphin-2.2-70b) for more details on the model.
+## Use with mlx
+
+```bash
+pip install mlx-lm
+```
+
+```python
+from mlx_lm import load, generate
+
+model, tokenizer = load("mlx-community/dolphin-2.2-70b")
+response = generate(model, tokenizer, prompt="hello", verbose=True)
+```

added_tokens.json

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

config.json

@@ -0,0 +1,81 @@
+{
+    "add_cross_attention": false,
+    "architectures": [
+        "LlamaForCausalLM"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "bad_words_ids": null,
+    "begin_suppress_tokens": null,
+    "bos_token_id": 1,
+    "chunk_size_feed_forward": 0,
+    "cross_attention_hidden_size": null,
+    "decoder_start_token_id": null,
+    "diversity_penalty": 0.0,
+    "do_sample": false,
+    "early_stopping": false,
+    "encoder_no_repeat_ngram_size": 0,
+    "eos_token_id": 32000,
+    "exponential_decay_length_penalty": null,
+    "finetuning_task": null,
+    "forced_bos_token_id": null,
+    "forced_eos_token_id": null,
+    "hidden_act": "silu",
+    "hidden_size": 8192,
+    "id2label": {
+        "0": "LABEL_0",
+        "1": "LABEL_1"
+    },
+    "initializer_range": 0.02,
+    "intermediate_size": 28672,
+    "is_decoder": false,
+    "is_encoder_decoder": false,
+    "label2id": {
+        "LABEL_0": 0,
+        "LABEL_1": 1
+    },
+    "length_penalty": 1.0,
+    "max_length": 20,
+    "max_position_embeddings": 4096,
+    "min_length": 0,
+    "model_type": "llama",
+    "no_repeat_ngram_size": 0,
+    "num_attention_heads": 64,
+    "num_beam_groups": 1,
+    "num_beams": 1,
+    "num_hidden_layers": 80,
+    "num_key_value_heads": 8,
+    "num_return_sequences": 1,
+    "output_attentions": false,
+    "output_hidden_states": false,
+    "output_scores": false,
+    "pad_token_id": 0,
+    "prefix": null,
+    "pretraining_tp": 1,
+    "problem_type": null,
+    "pruned_heads": {},
+    "remove_invalid_values": false,
+    "repetition_penalty": 1.0,
+    "return_dict": true,
+    "return_dict_in_generate": false,
+    "rms_norm_eps": 1e-05,
+    "rope_scaling": null,
+    "rope_theta": 10000.0,
+    "sep_token_id": null,
+    "suppress_tokens": null,
+    "task_specific_params": null,
+    "temperature": 1.0,
+    "tf_legacy_loss": false,
+    "tie_encoder_decoder": false,
+    "tie_word_embeddings": false,
+    "tokenizer_class": null,
+    "top_k": 50,
+    "top_p": 1.0,
+    "torch_dtype": "float16",
+    "torchscript": false,
+    "transformers_version": "4.39.1",
+    "typical_p": 1.0,
+    "use_bfloat16": false,
+    "use_cache": true,
+    "vocab_size": 32002
+}

model-00001-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48cbad61bb8b8d672ff54db2d85be8f0383e878c185739ac2b83f1cd13285aa7
+size 5188455242

model-00002-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e72e0ad916e0bd0405778f2a039230245fd955ceebeda6f81f62ed73cd1cdb3
+size 5133929517

model-00003-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce6b7a34f8bd1c71f75c4468153a11a393792e99db97df543acdb4bfaf97b7ae
+size 5133929525

model-00004-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:439fdb3f343dedb5085af47665e268a5464f35eabba353a6630a397bfa56097f
+size 5133929534

model-00005-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4bcb800836477746bb168c761a9a6336e42450851f763b018ab2c3b7b848ea2
+size 5133929554

model-00006-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44ecb031f0aaf6e2c40b161f03725112162372b17246afc9e36b663f53972ae8
+size 5133929546

model-00007-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:136c3ce49a6e56a08ee2cf21446bfd6c2f0a0b931b0f9915fc50dd37ca66ed4f
+size 5133929554

model-00008-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:469bb2b1d4e0151c65124283b3580f1169b1fcd5ba73680f6a462409d286b1d6
+size 5133929552

model-00009-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:804758bb84c8789cf47a569c2fae33e457215361edf7e9f0e119fb68eb904c02
+size 5133929550

model-00010-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b0561c458c51a9ae7c0a3b37231152a4202b6a7c33d00590325a7ea9ad83bd0
+size 5133929546

model-00011-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec9a0e3ba87cd151100c661257de6c3581230daf07277b858128ccfe3f93d104
+size 5133929548

model-00012-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f37d9a6d7293aab90a3c260a22dc367f2914ca1a43821b75281576d231f9ad07
+size 5133929554

model-00013-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95b492388ead98cda877b2876ed8e24eecc573d6c8d12bef7a54b18356d70550
+size 5133929554

model-00014-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42e0f5f139181ab4f842b380ff66bc1cb407fa107ed9caf8e2ec679048db04eb
+size 5133929534

model-00015-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a87389628ae27472bab84c5926fd1675d4459a02d0847422c6c7000a67fa5cee
+size 5133929552

model-00016-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34d9ec28612c6181c0696b2f61e45a5025cb1999e17d9b7614020623698a8db1
+size 5133929552

model-00017-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09feff62c797d446a45a21faeec070a7b8c62c2cfc85b14cd988f8febe4b9f5a
+size 5133929554

model-00018-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34000905da9366e46d0263bcb9f58ecce3353925d29214bcaa6ef0cc899bcdbb
+size 5133929544

model-00019-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:520da090d5cbecc3158f9a84af93df9b81efbaf4d57d5cbf198b7b7008919140
+size 5133929550

model-00020-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b935c44b488f65c0cd8d21b3a6b64fa854aeb13dbe1c7f1b5bc6faa057ea51b6
+size 5133929550

model-00021-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82d2c75526914eef177766d984f9f68e77cfb6b981edc53748f84822f75cb1ce
+size 5133929546

model-00022-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba89472d85324f5bb0b162aaa820c9baba496438e4e6e796d814a23694c9dad4
+size 5133929540

model-00023-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cad9a832914fc4eb3a7134fea779d97a610f234c7e37038b8efb1e4e62b8fb48
+size 5133929554

model-00024-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e492019a8d32d0f8c87dc2ba0b8c1d4dddcb9f425bc0244e6a4c82d1b296d12d
+size 5133929550

model-00025-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e040283fd16e0273afde974be01ee70eb339d0f4280ef4b8d4bab2c6a2822b1d
+size 5133929552

model-00026-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fefbb98961653f9660c9b1a7ae036f2fecef0f3f05ccf9851727c471536254fb
+size 5133929554

model-00027-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69e8fedebc45ef23da880216b5cf0de38cbf724802d28e0525fa267fd3ba53ae
+size 4416752195

model.safetensors.index.json

@@ -0,0 +1,730 @@
+{
+    "metadata": {
+        "total_size": 137953361920
+    },
+    "weight_map": {
+        "lm_head.weight": "model-00027-of-00027.safetensors",
+        "model.embed_tokens.weight": "model-00001-of-00027.safetensors",
+        "model.layers.0.input_layernorm.weight": "model-00001-of-00027.safetensors",
+        "model.layers.0.mlp.down_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.0.mlp.up_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00027.safetensors",
+        "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.1.input_layernorm.weight": "model-00001-of-00027.safetensors",
+        "model.layers.1.mlp.down_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.1.mlp.up_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00027.safetensors",
+        "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.10.input_layernorm.weight": "model-00004-of-00027.safetensors",
+        "model.layers.10.mlp.down_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.10.mlp.gate_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.10.mlp.up_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.10.post_attention_layernorm.weight": "model-00004-of-00027.safetensors",
+        "model.layers.10.self_attn.k_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.10.self_attn.o_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.10.self_attn.q_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.10.self_attn.v_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.11.input_layernorm.weight": "model-00005-of-00027.safetensors",
+        "model.layers.11.mlp.down_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.11.mlp.gate_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.11.mlp.up_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.11.post_attention_layernorm.weight": "model-00005-of-00027.safetensors",
+        "model.layers.11.self_attn.k_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.11.self_attn.o_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.11.self_attn.q_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.11.self_attn.v_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.12.input_layernorm.weight": "model-00005-of-00027.safetensors",
+        "model.layers.12.mlp.down_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.12.mlp.gate_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.12.mlp.up_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.12.post_attention_layernorm.weight": "model-00005-of-00027.safetensors",
+        "model.layers.12.self_attn.k_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.12.self_attn.o_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.12.self_attn.q_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.12.self_attn.v_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.13.input_layernorm.weight": "model-00005-of-00027.safetensors",
+        "model.layers.13.mlp.down_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.13.mlp.gate_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.13.mlp.up_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.13.post_attention_layernorm.weight": "model-00005-of-00027.safetensors",
+        "model.layers.13.self_attn.k_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.13.self_attn.o_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.13.self_attn.q_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.13.self_attn.v_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.14.input_layernorm.weight": "model-00006-of-00027.safetensors",
+        "model.layers.14.mlp.down_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.14.mlp.gate_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.14.mlp.up_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.14.post_attention_layernorm.weight": "model-00006-of-00027.safetensors",
+        "model.layers.14.self_attn.k_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.14.self_attn.o_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.14.self_attn.q_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.14.self_attn.v_proj.weight": "model-00005-of-00027.safetensors",
+        "model.layers.15.input_layernorm.weight": "model-00006-of-00027.safetensors",
+        "model.layers.15.mlp.down_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.15.mlp.gate_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.15.mlp.up_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.15.post_attention_layernorm.weight": "model-00006-of-00027.safetensors",
+        "model.layers.15.self_attn.k_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.15.self_attn.o_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.15.self_attn.q_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.15.self_attn.v_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.16.input_layernorm.weight": "model-00006-of-00027.safetensors",
+        "model.layers.16.mlp.down_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.16.mlp.gate_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.16.mlp.up_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.16.post_attention_layernorm.weight": "model-00006-of-00027.safetensors",
+        "model.layers.16.self_attn.k_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.16.self_attn.o_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.16.self_attn.q_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.16.self_attn.v_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.17.input_layernorm.weight": "model-00007-of-00027.safetensors",
+        "model.layers.17.mlp.down_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.17.mlp.gate_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.17.mlp.up_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.17.post_attention_layernorm.weight": "model-00007-of-00027.safetensors",
+        "model.layers.17.self_attn.k_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.17.self_attn.o_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.17.self_attn.q_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.17.self_attn.v_proj.weight": "model-00006-of-00027.safetensors",
+        "model.layers.18.input_layernorm.weight": "model-00007-of-00027.safetensors",
+        "model.layers.18.mlp.down_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.18.mlp.gate_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.18.mlp.up_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.18.post_attention_layernorm.weight": "model-00007-of-00027.safetensors",
+        "model.layers.18.self_attn.k_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.18.self_attn.o_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.18.self_attn.q_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.18.self_attn.v_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.19.input_layernorm.weight": "model-00007-of-00027.safetensors",
+        "model.layers.19.mlp.down_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.19.mlp.gate_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.19.mlp.up_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.19.post_attention_layernorm.weight": "model-00007-of-00027.safetensors",
+        "model.layers.19.self_attn.k_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.19.self_attn.o_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.19.self_attn.q_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.19.self_attn.v_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.2.input_layernorm.weight": "model-00002-of-00027.safetensors",
+        "model.layers.2.mlp.down_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.2.mlp.up_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00027.safetensors",
+        "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00027.safetensors",
+        "model.layers.20.input_layernorm.weight": "model-00008-of-00027.safetensors",
+        "model.layers.20.mlp.down_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.20.mlp.gate_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.20.mlp.up_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.20.post_attention_layernorm.weight": "model-00008-of-00027.safetensors",
+        "model.layers.20.self_attn.k_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.20.self_attn.o_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.20.self_attn.q_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.20.self_attn.v_proj.weight": "model-00007-of-00027.safetensors",
+        "model.layers.21.input_layernorm.weight": "model-00008-of-00027.safetensors",
+        "model.layers.21.mlp.down_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.21.mlp.gate_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.21.mlp.up_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.21.post_attention_layernorm.weight": "model-00008-of-00027.safetensors",
+        "model.layers.21.self_attn.k_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.21.self_attn.o_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.21.self_attn.q_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.21.self_attn.v_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.22.input_layernorm.weight": "model-00008-of-00027.safetensors",
+        "model.layers.22.mlp.down_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.22.mlp.gate_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.22.mlp.up_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.22.post_attention_layernorm.weight": "model-00008-of-00027.safetensors",
+        "model.layers.22.self_attn.k_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.22.self_attn.o_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.22.self_attn.q_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.22.self_attn.v_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.23.input_layernorm.weight": "model-00009-of-00027.safetensors",
+        "model.layers.23.mlp.down_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.23.mlp.gate_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.23.mlp.up_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.23.post_attention_layernorm.weight": "model-00009-of-00027.safetensors",
+        "model.layers.23.self_attn.k_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.23.self_attn.o_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.23.self_attn.q_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.23.self_attn.v_proj.weight": "model-00008-of-00027.safetensors",
+        "model.layers.24.input_layernorm.weight": "model-00009-of-00027.safetensors",
+        "model.layers.24.mlp.down_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.24.mlp.gate_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.24.mlp.up_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.24.post_attention_layernorm.weight": "model-00009-of-00027.safetensors",
+        "model.layers.24.self_attn.k_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.24.self_attn.o_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.24.self_attn.q_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.24.self_attn.v_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.25.input_layernorm.weight": "model-00009-of-00027.safetensors",
+        "model.layers.25.mlp.down_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.25.mlp.gate_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.25.mlp.up_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.25.post_attention_layernorm.weight": "model-00009-of-00027.safetensors",
+        "model.layers.25.self_attn.k_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.25.self_attn.o_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.25.self_attn.q_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.25.self_attn.v_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.26.input_layernorm.weight": "model-00010-of-00027.safetensors",
+        "model.layers.26.mlp.down_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.26.mlp.gate_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.26.mlp.up_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.26.post_attention_layernorm.weight": "model-00010-of-00027.safetensors",
+        "model.layers.26.self_attn.k_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.26.self_attn.o_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.26.self_attn.q_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.26.self_attn.v_proj.weight": "model-00009-of-00027.safetensors",
+        "model.layers.27.input_layernorm.weight": "model-00010-of-00027.safetensors",
+        "model.layers.27.mlp.down_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.27.mlp.gate_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.27.mlp.up_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.27.post_attention_layernorm.weight": "model-00010-of-00027.safetensors",
+        "model.layers.27.self_attn.k_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.27.self_attn.o_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.27.self_attn.q_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.27.self_attn.v_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.28.input_layernorm.weight": "model-00010-of-00027.safetensors",
+        "model.layers.28.mlp.down_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.28.mlp.gate_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.28.mlp.up_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.28.post_attention_layernorm.weight": "model-00010-of-00027.safetensors",
+        "model.layers.28.self_attn.k_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.28.self_attn.o_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.28.self_attn.q_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.28.self_attn.v_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.29.input_layernorm.weight": "model-00011-of-00027.safetensors",
+        "model.layers.29.mlp.down_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.29.mlp.gate_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.29.mlp.up_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.29.post_attention_layernorm.weight": "model-00011-of-00027.safetensors",
+        "model.layers.29.self_attn.k_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.29.self_attn.o_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.29.self_attn.q_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.29.self_attn.v_proj.weight": "model-00010-of-00027.safetensors",
+        "model.layers.3.input_layernorm.weight": "model-00002-of-00027.safetensors",
+        "model.layers.3.mlp.down_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.3.mlp.up_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00027.safetensors",
+        "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.30.input_layernorm.weight": "model-00011-of-00027.safetensors",
+        "model.layers.30.mlp.down_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.30.mlp.gate_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.30.mlp.up_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.30.post_attention_layernorm.weight": "model-00011-of-00027.safetensors",
+        "model.layers.30.self_attn.k_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.30.self_attn.o_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.30.self_attn.q_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.30.self_attn.v_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.31.input_layernorm.weight": "model-00011-of-00027.safetensors",
+        "model.layers.31.mlp.down_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.31.mlp.gate_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.31.mlp.up_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.31.post_attention_layernorm.weight": "model-00011-of-00027.safetensors",
+        "model.layers.31.self_attn.k_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.31.self_attn.o_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.31.self_attn.q_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.31.self_attn.v_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.32.input_layernorm.weight": "model-00012-of-00027.safetensors",
+        "model.layers.32.mlp.down_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.32.mlp.gate_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.32.mlp.up_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.32.post_attention_layernorm.weight": "model-00012-of-00027.safetensors",
+        "model.layers.32.self_attn.k_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.32.self_attn.o_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.32.self_attn.q_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.32.self_attn.v_proj.weight": "model-00011-of-00027.safetensors",
+        "model.layers.33.input_layernorm.weight": "model-00012-of-00027.safetensors",
+        "model.layers.33.mlp.down_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.33.mlp.gate_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.33.mlp.up_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.33.post_attention_layernorm.weight": "model-00012-of-00027.safetensors",
+        "model.layers.33.self_attn.k_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.33.self_attn.o_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.33.self_attn.q_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.33.self_attn.v_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.34.input_layernorm.weight": "model-00012-of-00027.safetensors",
+        "model.layers.34.mlp.down_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.34.mlp.gate_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.34.mlp.up_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.34.post_attention_layernorm.weight": "model-00012-of-00027.safetensors",
+        "model.layers.34.self_attn.k_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.34.self_attn.o_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.34.self_attn.q_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.34.self_attn.v_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.35.input_layernorm.weight": "model-00013-of-00027.safetensors",
+        "model.layers.35.mlp.down_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.35.mlp.gate_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.35.mlp.up_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.35.post_attention_layernorm.weight": "model-00013-of-00027.safetensors",
+        "model.layers.35.self_attn.k_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.35.self_attn.o_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.35.self_attn.q_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.35.self_attn.v_proj.weight": "model-00012-of-00027.safetensors",
+        "model.layers.36.input_layernorm.weight": "model-00013-of-00027.safetensors",
+        "model.layers.36.mlp.down_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.36.mlp.gate_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.36.mlp.up_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.36.post_attention_layernorm.weight": "model-00013-of-00027.safetensors",
+        "model.layers.36.self_attn.k_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.36.self_attn.o_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.36.self_attn.q_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.36.self_attn.v_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.37.input_layernorm.weight": "model-00013-of-00027.safetensors",
+        "model.layers.37.mlp.down_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.37.mlp.gate_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.37.mlp.up_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.37.post_attention_layernorm.weight": "model-00013-of-00027.safetensors",
+        "model.layers.37.self_attn.k_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.37.self_attn.o_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.37.self_attn.q_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.37.self_attn.v_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.38.input_layernorm.weight": "model-00014-of-00027.safetensors",
+        "model.layers.38.mlp.down_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.38.mlp.gate_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.38.mlp.up_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.38.post_attention_layernorm.weight": "model-00014-of-00027.safetensors",
+        "model.layers.38.self_attn.k_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.38.self_attn.o_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.38.self_attn.q_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.38.self_attn.v_proj.weight": "model-00013-of-00027.safetensors",
+        "model.layers.39.input_layernorm.weight": "model-00014-of-00027.safetensors",
+        "model.layers.39.mlp.down_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.39.mlp.gate_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.39.mlp.up_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.39.post_attention_layernorm.weight": "model-00014-of-00027.safetensors",
+        "model.layers.39.self_attn.k_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.39.self_attn.o_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.39.self_attn.q_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.39.self_attn.v_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.4.input_layernorm.weight": "model-00002-of-00027.safetensors",
+        "model.layers.4.mlp.down_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.4.mlp.up_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00027.safetensors",
+        "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.40.input_layernorm.weight": "model-00014-of-00027.safetensors",
+        "model.layers.40.mlp.down_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.40.mlp.gate_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.40.mlp.up_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.40.post_attention_layernorm.weight": "model-00014-of-00027.safetensors",
+        "model.layers.40.self_attn.k_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.40.self_attn.o_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.40.self_attn.q_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.40.self_attn.v_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.41.input_layernorm.weight": "model-00015-of-00027.safetensors",
+        "model.layers.41.mlp.down_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.41.mlp.gate_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.41.mlp.up_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.41.post_attention_layernorm.weight": "model-00015-of-00027.safetensors",
+        "model.layers.41.self_attn.k_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.41.self_attn.o_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.41.self_attn.q_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.41.self_attn.v_proj.weight": "model-00014-of-00027.safetensors",
+        "model.layers.42.input_layernorm.weight": "model-00015-of-00027.safetensors",
+        "model.layers.42.mlp.down_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.42.mlp.gate_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.42.mlp.up_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.42.post_attention_layernorm.weight": "model-00015-of-00027.safetensors",
+        "model.layers.42.self_attn.k_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.42.self_attn.o_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.42.self_attn.q_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.42.self_attn.v_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.43.input_layernorm.weight": "model-00015-of-00027.safetensors",
+        "model.layers.43.mlp.down_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.43.mlp.gate_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.43.mlp.up_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.43.post_attention_layernorm.weight": "model-00015-of-00027.safetensors",
+        "model.layers.43.self_attn.k_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.43.self_attn.o_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.43.self_attn.q_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.43.self_attn.v_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.44.input_layernorm.weight": "model-00016-of-00027.safetensors",
+        "model.layers.44.mlp.down_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.44.mlp.gate_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.44.mlp.up_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.44.post_attention_layernorm.weight": "model-00016-of-00027.safetensors",
+        "model.layers.44.self_attn.k_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.44.self_attn.o_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.44.self_attn.q_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.44.self_attn.v_proj.weight": "model-00015-of-00027.safetensors",
+        "model.layers.45.input_layernorm.weight": "model-00016-of-00027.safetensors",
+        "model.layers.45.mlp.down_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.45.mlp.gate_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.45.mlp.up_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.45.post_attention_layernorm.weight": "model-00016-of-00027.safetensors",
+        "model.layers.45.self_attn.k_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.45.self_attn.o_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.45.self_attn.q_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.45.self_attn.v_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.46.input_layernorm.weight": "model-00016-of-00027.safetensors",
+        "model.layers.46.mlp.down_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.46.mlp.gate_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.46.mlp.up_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.46.post_attention_layernorm.weight": "model-00016-of-00027.safetensors",
+        "model.layers.46.self_attn.k_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.46.self_attn.o_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.46.self_attn.q_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.46.self_attn.v_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.47.input_layernorm.weight": "model-00017-of-00027.safetensors",
+        "model.layers.47.mlp.down_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.47.mlp.gate_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.47.mlp.up_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.47.post_attention_layernorm.weight": "model-00017-of-00027.safetensors",
+        "model.layers.47.self_attn.k_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.47.self_attn.o_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.47.self_attn.q_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.47.self_attn.v_proj.weight": "model-00016-of-00027.safetensors",
+        "model.layers.48.input_layernorm.weight": "model-00017-of-00027.safetensors",
+        "model.layers.48.mlp.down_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.48.mlp.gate_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.48.mlp.up_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.48.post_attention_layernorm.weight": "model-00017-of-00027.safetensors",
+        "model.layers.48.self_attn.k_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.48.self_attn.o_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.48.self_attn.q_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.48.self_attn.v_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.49.input_layernorm.weight": "model-00017-of-00027.safetensors",
+        "model.layers.49.mlp.down_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.49.mlp.gate_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.49.mlp.up_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.49.post_attention_layernorm.weight": "model-00017-of-00027.safetensors",
+        "model.layers.49.self_attn.k_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.49.self_attn.o_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.49.self_attn.q_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.49.self_attn.v_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.5.input_layernorm.weight": "model-00003-of-00027.safetensors",
+        "model.layers.5.mlp.down_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.5.mlp.up_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.5.post_attention_layernorm.weight": "model-00003-of-00027.safetensors",
+        "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00027.safetensors",
+        "model.layers.50.input_layernorm.weight": "model-00018-of-00027.safetensors",
+        "model.layers.50.mlp.down_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.50.mlp.gate_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.50.mlp.up_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.50.post_attention_layernorm.weight": "model-00018-of-00027.safetensors",
+        "model.layers.50.self_attn.k_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.50.self_attn.o_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.50.self_attn.q_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.50.self_attn.v_proj.weight": "model-00017-of-00027.safetensors",
+        "model.layers.51.input_layernorm.weight": "model-00018-of-00027.safetensors",
+        "model.layers.51.mlp.down_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.51.mlp.gate_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.51.mlp.up_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.51.post_attention_layernorm.weight": "model-00018-of-00027.safetensors",
+        "model.layers.51.self_attn.k_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.51.self_attn.o_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.51.self_attn.q_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.51.self_attn.v_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.52.input_layernorm.weight": "model-00018-of-00027.safetensors",
+        "model.layers.52.mlp.down_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.52.mlp.gate_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.52.mlp.up_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.52.post_attention_layernorm.weight": "model-00018-of-00027.safetensors",
+        "model.layers.52.self_attn.k_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.52.self_attn.o_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.52.self_attn.q_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.52.self_attn.v_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.53.input_layernorm.weight": "model-00019-of-00027.safetensors",
+        "model.layers.53.mlp.down_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.53.mlp.gate_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.53.mlp.up_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.53.post_attention_layernorm.weight": "model-00019-of-00027.safetensors",
+        "model.layers.53.self_attn.k_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.53.self_attn.o_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.53.self_attn.q_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.53.self_attn.v_proj.weight": "model-00018-of-00027.safetensors",
+        "model.layers.54.input_layernorm.weight": "model-00019-of-00027.safetensors",
+        "model.layers.54.mlp.down_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.54.mlp.gate_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.54.mlp.up_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.54.post_attention_layernorm.weight": "model-00019-of-00027.safetensors",
+        "model.layers.54.self_attn.k_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.54.self_attn.o_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.54.self_attn.q_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.54.self_attn.v_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.55.input_layernorm.weight": "model-00019-of-00027.safetensors",
+        "model.layers.55.mlp.down_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.55.mlp.gate_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.55.mlp.up_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.55.post_attention_layernorm.weight": "model-00019-of-00027.safetensors",
+        "model.layers.55.self_attn.k_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.55.self_attn.o_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.55.self_attn.q_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.55.self_attn.v_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.56.input_layernorm.weight": "model-00020-of-00027.safetensors",
+        "model.layers.56.mlp.down_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.56.mlp.gate_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.56.mlp.up_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.56.post_attention_layernorm.weight": "model-00020-of-00027.safetensors",
+        "model.layers.56.self_attn.k_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.56.self_attn.o_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.56.self_attn.q_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.56.self_attn.v_proj.weight": "model-00019-of-00027.safetensors",
+        "model.layers.57.input_layernorm.weight": "model-00020-of-00027.safetensors",
+        "model.layers.57.mlp.down_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.57.mlp.gate_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.57.mlp.up_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.57.post_attention_layernorm.weight": "model-00020-of-00027.safetensors",
+        "model.layers.57.self_attn.k_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.57.self_attn.o_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.57.self_attn.q_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.57.self_attn.v_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.58.input_layernorm.weight": "model-00020-of-00027.safetensors",
+        "model.layers.58.mlp.down_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.58.mlp.gate_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.58.mlp.up_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.58.post_attention_layernorm.weight": "model-00020-of-00027.safetensors",
+        "model.layers.58.self_attn.k_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.58.self_attn.o_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.58.self_attn.q_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.58.self_attn.v_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.59.input_layernorm.weight": "model-00021-of-00027.safetensors",
+        "model.layers.59.mlp.down_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.59.mlp.gate_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.59.mlp.up_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.59.post_attention_layernorm.weight": "model-00021-of-00027.safetensors",
+        "model.layers.59.self_attn.k_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.59.self_attn.o_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.59.self_attn.q_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.59.self_attn.v_proj.weight": "model-00020-of-00027.safetensors",
+        "model.layers.6.input_layernorm.weight": "model-00003-of-00027.safetensors",
+        "model.layers.6.mlp.down_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.6.mlp.gate_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.6.mlp.up_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.6.post_attention_layernorm.weight": "model-00003-of-00027.safetensors",
+        "model.layers.6.self_attn.k_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.6.self_attn.o_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.6.self_attn.q_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.6.self_attn.v_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.60.input_layernorm.weight": "model-00021-of-00027.safetensors",
+        "model.layers.60.mlp.down_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.60.mlp.gate_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.60.mlp.up_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.60.post_attention_layernorm.weight": "model-00021-of-00027.safetensors",
+        "model.layers.60.self_attn.k_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.60.self_attn.o_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.60.self_attn.q_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.60.self_attn.v_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.61.input_layernorm.weight": "model-00021-of-00027.safetensors",
+        "model.layers.61.mlp.down_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.61.mlp.gate_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.61.mlp.up_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.61.post_attention_layernorm.weight": "model-00021-of-00027.safetensors",
+        "model.layers.61.self_attn.k_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.61.self_attn.o_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.61.self_attn.q_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.61.self_attn.v_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.62.input_layernorm.weight": "model-00022-of-00027.safetensors",
+        "model.layers.62.mlp.down_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.62.mlp.gate_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.62.mlp.up_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.62.post_attention_layernorm.weight": "model-00022-of-00027.safetensors",
+        "model.layers.62.self_attn.k_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.62.self_attn.o_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.62.self_attn.q_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.62.self_attn.v_proj.weight": "model-00021-of-00027.safetensors",
+        "model.layers.63.input_layernorm.weight": "model-00022-of-00027.safetensors",
+        "model.layers.63.mlp.down_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.63.mlp.gate_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.63.mlp.up_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.63.post_attention_layernorm.weight": "model-00022-of-00027.safetensors",
+        "model.layers.63.self_attn.k_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.63.self_attn.o_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.63.self_attn.q_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.63.self_attn.v_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.64.input_layernorm.weight": "model-00022-of-00027.safetensors",
+        "model.layers.64.mlp.down_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.64.mlp.gate_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.64.mlp.up_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.64.post_attention_layernorm.weight": "model-00022-of-00027.safetensors",
+        "model.layers.64.self_attn.k_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.64.self_attn.o_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.64.self_attn.q_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.64.self_attn.v_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.65.input_layernorm.weight": "model-00023-of-00027.safetensors",
+        "model.layers.65.mlp.down_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.65.mlp.gate_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.65.mlp.up_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.65.post_attention_layernorm.weight": "model-00023-of-00027.safetensors",
+        "model.layers.65.self_attn.k_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.65.self_attn.o_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.65.self_attn.q_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.65.self_attn.v_proj.weight": "model-00022-of-00027.safetensors",
+        "model.layers.66.input_layernorm.weight": "model-00023-of-00027.safetensors",
+        "model.layers.66.mlp.down_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.66.mlp.gate_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.66.mlp.up_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.66.post_attention_layernorm.weight": "model-00023-of-00027.safetensors",
+        "model.layers.66.self_attn.k_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.66.self_attn.o_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.66.self_attn.q_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.66.self_attn.v_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.67.input_layernorm.weight": "model-00023-of-00027.safetensors",
+        "model.layers.67.mlp.down_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.67.mlp.gate_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.67.mlp.up_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.67.post_attention_layernorm.weight": "model-00023-of-00027.safetensors",
+        "model.layers.67.self_attn.k_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.67.self_attn.o_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.67.self_attn.q_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.67.self_attn.v_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.68.input_layernorm.weight": "model-00024-of-00027.safetensors",
+        "model.layers.68.mlp.down_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.68.mlp.gate_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.68.mlp.up_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.68.post_attention_layernorm.weight": "model-00024-of-00027.safetensors",
+        "model.layers.68.self_attn.k_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.68.self_attn.o_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.68.self_attn.q_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.68.self_attn.v_proj.weight": "model-00023-of-00027.safetensors",
+        "model.layers.69.input_layernorm.weight": "model-00024-of-00027.safetensors",
+        "model.layers.69.mlp.down_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.69.mlp.gate_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.69.mlp.up_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.69.post_attention_layernorm.weight": "model-00024-of-00027.safetensors",
+        "model.layers.69.self_attn.k_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.69.self_attn.o_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.69.self_attn.q_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.69.self_attn.v_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.7.input_layernorm.weight": "model-00003-of-00027.safetensors",
+        "model.layers.7.mlp.down_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.7.mlp.gate_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.7.mlp.up_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.7.post_attention_layernorm.weight": "model-00003-of-00027.safetensors",
+        "model.layers.7.self_attn.k_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.7.self_attn.o_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.7.self_attn.q_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.7.self_attn.v_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.70.input_layernorm.weight": "model-00024-of-00027.safetensors",
+        "model.layers.70.mlp.down_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.70.mlp.gate_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.70.mlp.up_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.70.post_attention_layernorm.weight": "model-00024-of-00027.safetensors",
+        "model.layers.70.self_attn.k_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.70.self_attn.o_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.70.self_attn.q_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.70.self_attn.v_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.71.input_layernorm.weight": "model-00025-of-00027.safetensors",
+        "model.layers.71.mlp.down_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.71.mlp.gate_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.71.mlp.up_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.71.post_attention_layernorm.weight": "model-00025-of-00027.safetensors",
+        "model.layers.71.self_attn.k_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.71.self_attn.o_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.71.self_attn.q_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.71.self_attn.v_proj.weight": "model-00024-of-00027.safetensors",
+        "model.layers.72.input_layernorm.weight": "model-00025-of-00027.safetensors",
+        "model.layers.72.mlp.down_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.72.mlp.gate_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.72.mlp.up_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.72.post_attention_layernorm.weight": "model-00025-of-00027.safetensors",
+        "model.layers.72.self_attn.k_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.72.self_attn.o_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.72.self_attn.q_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.72.self_attn.v_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.73.input_layernorm.weight": "model-00025-of-00027.safetensors",
+        "model.layers.73.mlp.down_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.73.mlp.gate_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.73.mlp.up_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.73.post_attention_layernorm.weight": "model-00025-of-00027.safetensors",
+        "model.layers.73.self_attn.k_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.73.self_attn.o_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.73.self_attn.q_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.73.self_attn.v_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.74.input_layernorm.weight": "model-00026-of-00027.safetensors",
+        "model.layers.74.mlp.down_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.74.mlp.gate_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.74.mlp.up_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.74.post_attention_layernorm.weight": "model-00026-of-00027.safetensors",
+        "model.layers.74.self_attn.k_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.74.self_attn.o_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.74.self_attn.q_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.74.self_attn.v_proj.weight": "model-00025-of-00027.safetensors",
+        "model.layers.75.input_layernorm.weight": "model-00026-of-00027.safetensors",
+        "model.layers.75.mlp.down_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.75.mlp.gate_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.75.mlp.up_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.75.post_attention_layernorm.weight": "model-00026-of-00027.safetensors",
+        "model.layers.75.self_attn.k_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.75.self_attn.o_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.75.self_attn.q_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.75.self_attn.v_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.76.input_layernorm.weight": "model-00026-of-00027.safetensors",
+        "model.layers.76.mlp.down_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.76.mlp.gate_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.76.mlp.up_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.76.post_attention_layernorm.weight": "model-00026-of-00027.safetensors",
+        "model.layers.76.self_attn.k_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.76.self_attn.o_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.76.self_attn.q_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.76.self_attn.v_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.77.input_layernorm.weight": "model-00027-of-00027.safetensors",
+        "model.layers.77.mlp.down_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.77.mlp.gate_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.77.mlp.up_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.77.post_attention_layernorm.weight": "model-00027-of-00027.safetensors",
+        "model.layers.77.self_attn.k_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.77.self_attn.o_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.77.self_attn.q_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.77.self_attn.v_proj.weight": "model-00026-of-00027.safetensors",
+        "model.layers.78.input_layernorm.weight": "model-00027-of-00027.safetensors",
+        "model.layers.78.mlp.down_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.78.mlp.gate_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.78.mlp.up_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.78.post_attention_layernorm.weight": "model-00027-of-00027.safetensors",
+        "model.layers.78.self_attn.k_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.78.self_attn.o_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.78.self_attn.q_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.78.self_attn.v_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.79.input_layernorm.weight": "model-00027-of-00027.safetensors",
+        "model.layers.79.mlp.down_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.79.mlp.gate_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.79.mlp.up_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.79.post_attention_layernorm.weight": "model-00027-of-00027.safetensors",
+        "model.layers.79.self_attn.k_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.79.self_attn.o_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.79.self_attn.q_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.79.self_attn.v_proj.weight": "model-00027-of-00027.safetensors",
+        "model.layers.8.input_layernorm.weight": "model-00004-of-00027.safetensors",
+        "model.layers.8.mlp.down_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.8.mlp.gate_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.8.mlp.up_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.8.post_attention_layernorm.weight": "model-00004-of-00027.safetensors",
+        "model.layers.8.self_attn.k_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.8.self_attn.o_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.8.self_attn.q_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.8.self_attn.v_proj.weight": "model-00003-of-00027.safetensors",
+        "model.layers.9.input_layernorm.weight": "model-00004-of-00027.safetensors",
+        "model.layers.9.mlp.down_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.9.mlp.gate_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.9.mlp.up_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.9.post_attention_layernorm.weight": "model-00004-of-00027.safetensors",
+        "model.layers.9.self_attn.k_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.9.self_attn.o_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.9.self_attn.q_proj.weight": "model-00004-of-00027.safetensors",
+        "model.layers.9.self_attn.v_proj.weight": "model-00004-of-00027.safetensors",
+        "model.norm.weight": "model-00027-of-00027.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 sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,66 @@
+{
+  "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
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|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": false,
+  "max_length": 2048,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "stride": 0,
+  "tokenizer_class": "LlamaTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "trust_remote_code": false,
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true,
+  "use_fast": true
+}