初始化项目，由ModelHub XC社区提供模型

Model: chargoddard/llama2-22b
Source: Original Platform

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,27 @@
+---
+model_type: llama
+pipeline_tag: text-generation
+datasets:
+- togethercomputer/RedPajama-Data-1T-Sample
+tags:
+- llama
+---
+
+This is [Llama 2 13b](https://huggingface.co/meta-llama/Llama-2-13b-hf) with some additional attention heads from original-flavor Llama 33b frankensteined on.
+
+Fine-tuned on ~10M tokens from RedPajama to settle in the transplants a little.
+
+Not intended for use as-is - this model is meant to serve as a base for further tuning, hopefully with a greater capacity for learning than 13b.
+# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
+Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_chargoddard__llama2-22b)
+
+| Metric                | Value                     |
+|-----------------------|---------------------------|
+| Avg.                  | 46.85   |
+| ARC (25-shot)         | 58.53          |
+| HellaSwag (10-shot)   | 82.55    |
+| MMLU (5-shot)         | 54.68         |
+| TruthfulQA (0-shot)   | 39.84   |
+| Winogrande (5-shot)   | 76.32   |
+| GSM8K (5-shot)        | 9.93        |
+| DROP (3-shot)         | 6.08         |

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "/root/llama2-22b/",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 6656,
+  "initializer_range": 0.02,
+  "intermediate_size": 17920,
+  "max_position_embeddings": 4098,
+  "model_type": "llama",
+  "num_attention_heads": 52,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 52,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.32.0.dev0",
+  "use_cache": false,
+  "vocab_size": 32000
+}

frankenllama_22b.py

@@ -0,0 +1,188 @@
+#!/usr/bin/env python3
+# Charles O. Goddard
+# 7/20/2023
+"""Script used to generate the base frankenmerge. Output will need fine-tuning to be useful."""
+
+import copy
+import torch
+from torch import Tensor, nn
+import transformers
+
+from transformers.models.llama.modeling_llama import (
+    LlamaForCausalLM,
+    LlamaDecoderLayer,
+)
+from transformers import LlamaForCausalLM, LlamaConfig
+
+import torch
+import transformers
+import numpy as np
+
+
+MODEL_NAME_13B = "meta-llama/Llama-2-13b-hf"  # primary model
+MODEL_NAME_33B = "huggyllama/llama-30b"  # donor
+BLOCK_DIAGONAL = True
+# If BLOCK_DIAGONAL is set to True, each tensor in the resultant model will form a
+# block diagonal matrix, as illustrated below:
+
+# a a a 0 0
+# a a a 0 0
+# a a a 0 0
+# 0 0 0 b b
+# 0 0 0 b b
+
+# In this configuration, the states (hidden and intermediate) from the original 
+# and donor models are completely decoupled. That is, the hidden states
+# corresponding to the original model remain unchanged, and the new dimensions 
+# added from the donor model do not depend on the hidden states of the original model.
+
+# If BLOCK_DIAGONAL is set to False, the tensors will instead have the following form:
+
+# a a a 0 0
+# a a a 0 0
+# a a a 0 0
+# b b b b b
+# b b b b b
+
+# In this case, the output of the newly added attention heads depends on the hidden 
+# state values as if they were part of the donor model. Although the original model's
+# hidden states remain unchanged in either case, interaction between the new and old
+# features will result in features of varying usefulness.
+
+
+class NoInit:
+    def __enter__(self):
+        def noop(*args, **kwargs):
+            pass
+
+        (k, u, n) = (
+            torch.nn.init.kaiming_uniform_,
+            torch.nn.init.uniform_,
+            torch.nn.init.normal_,
+        )
+        torch.nn.init.kaiming_uniform_ = noop
+        torch.nn.init.uniform_ = noop
+        torch.nn.init.normal_ = noop
+
+        transformers.modeling_utils._init_weights = False
+        self.funcs = (k, u, n)
+
+    def __exit__(self, *args):
+        (k, u, n) = self.funcs
+        (
+            torch.nn.init.kaiming_uniform_,
+            torch.nn.init.uniform_,
+            torch.nn.init.normal_,
+        ) = (
+            k,
+            u,
+            n,
+        )
+        transformers.modeling_utils._init_weights = True
+
+
+def format_kmb(n, digits=None):
+    n = int(n)
+    if n < 1000:
+        return str(n)
+    elif n < 1000_000:
+        return f"{round(n/1000, digits)}k"
+    elif n < 1000 * 1000 * 1000:
+        return f"{round(n/(1000*1000), digits)}m"
+    else:
+        return f"{round(n/(1000*1000*1000), digits)}b"
+
+
+def count_params(model):
+    model_parameters = filter(lambda p: p.requires_grad, model.parameters())
+    params = sum([np.prod(p.size()) for p in model_parameters])
+    return int(params)
+
+
+torch.set_default_dtype(torch.float16)
+
+config_13b: LlamaConfig = LlamaConfig.from_pretrained(MODEL_NAME_13B)
+config_33b: LlamaConfig = LlamaConfig.from_pretrained(MODEL_NAME_33B)
+config_more = copy.deepcopy(config_13b)
+config_more.intermediate_size = config_33b.intermediate_size
+config_more.hidden_size = config_33b.hidden_size
+config_more.num_key_value_heads = config_33b.num_key_value_heads
+config_more.num_attention_heads = config_33b.num_key_value_heads
+
+print(config_more)
+
+with NoInit():
+    model = LlamaForCausalLM(config_more)
+
+print(f"{format_kmb(count_params(model), 3)} parameters")
+
+
+def merge_tensors_inplace(dest: Tensor, s0: Tensor, s1: Tensor, block_diagonal: bool):
+    dest.zero_()
+    if block_diagonal:
+        dest[s0.shape[0] :, s0.shape[1] :] = s1[
+            s0.shape[0] : dest.shape[0],
+            s0.shape[1] : dest.shape[1],
+        ]
+    else:
+        dest[s0.shape[0] :, :] = s1[
+            s0.shape[0] : dest.shape[0],
+            : dest.shape[1],
+        ]
+    dest[: s0.shape[0], : s0.shape[1]] = s0
+
+
+with NoInit():
+    donor_13b = (
+        LlamaForCausalLM.from_pretrained(MODEL_NAME_13B).to(torch.float16).eval()
+    )
+    donor_33b = (
+        LlamaForCausalLM.from_pretrained(MODEL_NAME_33B).to(torch.float16).eval()
+    )
+
+with torch.no_grad():
+    for layer_idx in range(len(model.model.layers)):
+        layer: LlamaDecoderLayer = model.model.layers[layer_idx]
+        l13: LlamaDecoderLayer = donor_13b.model.layers[layer_idx]
+        l33: LlamaDecoderLayer = donor_33b.model.layers[layer_idx]
+
+        for name in ("q_proj", "k_proj", "v_proj", "o_proj"):
+            dest: nn.Linear = getattr(layer.self_attn, name)
+            s13: nn.Linear = getattr(l13.self_attn, name)
+            s33: nn.Linear = getattr(l33.self_attn, name)
+            merge_tensors_inplace(dest.weight, s13.weight, s33.weight, BLOCK_DIAGONAL)
+
+        for name in ("up_proj", "gate_proj", "down_proj"):
+            dest: nn.Linear = getattr(layer.mlp, name)
+            s13: nn.Linear = getattr(l13.mlp, name)
+            s33: nn.Linear = getattr(l33.mlp, name)
+            merge_tensors_inplace(dest.weight, s13.weight, s33.weight, BLOCK_DIAGONAL)
+
+        layer.input_layernorm.weight[:] = l33.input_layernorm.weight[
+            : layer.input_layernorm.weight.shape[0]
+        ]
+        layer.input_layernorm.weight[
+            : l13.input_layernorm.weight.shape[0]
+        ] = l13.input_layernorm.weight
+        layer.post_attention_layernorm.weight[:] = l33.post_attention_layernorm.weight[
+            : layer.post_attention_layernorm.weight.shape[0]
+        ]
+        layer.post_attention_layernorm.weight[
+            : l13.post_attention_layernorm.weight.shape[0]
+        ] = l13.post_attention_layernorm.weight
+
+    # have initial output depend on only original llama2-13b features, so model
+    # starts unimpaired and can learn to incorporate the new features as well
+    model.lm_head.weight.zero_()
+    model.lm_head.weight[
+        : donor_13b.lm_head.weight.shape[0], : donor_13b.lm_head.weight.shape[1]
+    ] = donor_13b.lm_head.weight
+
+    merge_tensors_inplace(
+        model.model.embed_tokens.weight,
+        donor_13b.model.embed_tokens.weight,
+        donor_33b.model.embed_tokens.weight,
+        BLOCK_DIAGONAL,
+    )
+
+model.save_pretrained("./llama2-22b/", safe_serialization=True)

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "pad_token_id": 0,
+  "transformers_version": "4.32.0.dev0"
+}

pytorch_model-00001-of-00005.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70019781140edfa10f01730a1e20b649a82cd104981bbd42eb175c985c581741
+size 9818324691

pytorch_model-00002-of-00005.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d96b90f05e26d973b86c512c33b121683c30efd88eb6ee2bca08fae7a128366
+size 9958102871

pytorch_model-00003-of-00005.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd2b8673c387c5497edbeee036c9fb6072f0101deb7fb97158fc53cb65595ed0
+size 9896734715

pytorch_model-00004-of-00005.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b60d983a87dd4f8450583eedda634dc452152a719ea6ce86bbd77548dd82240d
+size 9869470609

pytorch_model-00005-of-00005.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d950dcd51ecfdc5cc77c357acb2dd2766cbda8c538ab6e333f62abef67feac0c
+size 4113434135

pytorch_model.bin.index.json

@@ -0,0 +1,410 @@
+{
+  "metadata": {
+    "total_size": 43655923712
+  },
+  "weight_map": {
+    "lm_head.weight": "pytorch_model-00005-of-00005.bin",
+    "model.embed_tokens.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.0.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00005.bin",
+    "model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.1.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00005.bin",
+    "model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.10.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.10.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.10.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.10.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.10.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.10.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.10.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00005.bin",
+    "model.layers.10.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.11.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.11.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.11.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.11.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.11.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.11.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.11.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00005.bin",
+    "model.layers.11.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.12.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.12.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.12.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.12.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.12.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.12.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.12.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00005.bin",
+    "model.layers.12.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.13.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.13.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.13.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.13.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.13.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.13.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.13.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00005.bin",
+    "model.layers.13.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.14.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.14.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.14.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.14.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.14.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.14.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.14.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00005.bin",
+    "model.layers.14.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.15.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.15.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.15.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.15.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.15.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.15.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.15.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00005.bin",
+    "model.layers.15.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.16.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.16.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.16.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.16.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.16.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.16.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.16.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00005.bin",
+    "model.layers.16.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.17.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.17.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.17.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.17.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.17.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.17.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.17.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00005.bin",
+    "model.layers.17.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.18.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.18.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.18.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.18.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.18.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.18.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.18.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00005.bin",
+    "model.layers.18.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.19.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.19.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.19.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.19.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.19.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.19.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.19.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00005.bin",
+    "model.layers.19.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.2.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00005.bin",
+    "model.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.20.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.20.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.20.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.20.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.20.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.20.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.20.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00005.bin",
+    "model.layers.20.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.21.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.21.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.21.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.21.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.21.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.21.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.21.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00005.bin",
+    "model.layers.21.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.22.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.22.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.22.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.22.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.22.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.22.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.22.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00005.bin",
+    "model.layers.22.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.23.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.23.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.23.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.23.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.23.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.23.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.23.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00005.bin",
+    "model.layers.23.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.24.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.24.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.24.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.24.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.24.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.24.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.24.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00005.bin",
+    "model.layers.24.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.25.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.25.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.25.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.25.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.25.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.25.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.25.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00005.bin",
+    "model.layers.25.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.26.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.26.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.26.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.26.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.26.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.26.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.26.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00005.bin",
+    "model.layers.26.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.27.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.27.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.27.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.27.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.27.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.27.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.27.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00005.bin",
+    "model.layers.27.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
+    "model.layers.28.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.28.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.28.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.28.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.28.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.28.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.28.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00005.bin",
+    "model.layers.28.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.29.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.29.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.29.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.29.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.29.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.29.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.29.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00005.bin",
+    "model.layers.29.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.3.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00005.bin",
+    "model.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.30.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.30.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.30.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.30.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.30.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.30.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.30.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00005.bin",
+    "model.layers.30.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.31.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.31.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.31.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.31.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.31.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.31.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.31.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00005.bin",
+    "model.layers.31.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.32.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.32.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.32.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.32.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.32.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.32.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.32.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.32.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.32.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00005.bin",
+    "model.layers.32.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.33.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.33.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.33.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.33.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.33.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.33.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.33.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.33.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.33.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00005.bin",
+    "model.layers.33.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.34.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.34.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.34.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.34.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.34.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.34.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.34.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.34.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.34.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00005.bin",
+    "model.layers.34.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.35.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.35.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.35.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.35.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.35.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.35.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.35.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.35.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.35.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00005.bin",
+    "model.layers.35.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.36.input_layernorm.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.36.mlp.down_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.36.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.36.mlp.up_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.36.post_attention_layernorm.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.36.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.36.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.36.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.36.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00005.bin",
+    "model.layers.36.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
+    "model.layers.37.input_layernorm.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.37.mlp.down_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.37.mlp.gate_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.37.mlp.up_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.37.post_attention_layernorm.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.37.self_attn.k_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.37.self_attn.o_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.37.self_attn.q_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.37.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00005.bin",
+    "model.layers.37.self_attn.v_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.38.input_layernorm.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.38.mlp.down_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.38.mlp.gate_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.38.mlp.up_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.38.post_attention_layernorm.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.38.self_attn.k_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.38.self_attn.o_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.38.self_attn.q_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.38.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00005.bin",
+    "model.layers.38.self_attn.v_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.39.input_layernorm.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.39.mlp.down_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.39.mlp.gate_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.39.mlp.up_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.39.post_attention_layernorm.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.39.self_attn.k_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.39.self_attn.o_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.39.self_attn.q_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.39.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00005.bin",
+    "model.layers.39.self_attn.v_proj.weight": "pytorch_model-00005-of-00005.bin",
+    "model.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.4.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00005.bin",
+    "model.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.5.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.5.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.5.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00005.bin",
+    "model.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.6.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.6.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.6.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.6.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.6.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.6.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.6.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00005.bin",
+    "model.layers.6.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.7.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.7.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.7.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.7.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.7.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.7.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.7.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00005.bin",
+    "model.layers.7.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.8.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.8.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.8.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.8.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.8.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.8.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.8.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00005.bin",
+    "model.layers.8.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
+    "model.layers.9.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.9.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.9.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.9.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.9.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.9.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.layers.9.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00005.bin",
+    "model.layers.9.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
+    "model.norm.weight": "pytorch_model-00005-of-00005.bin"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "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": "<unk>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,34 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "bos_token": {
+    "__type": "AddedToken",
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "clean_up_tokenization_spaces": false,
+  "eos_token": {
+    "__type": "AddedToken",
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": {
+    "__type": "AddedToken",
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}