upload model files

README.md

@@ -1,3 +1,31 @@
 ---
 license: apache-2.0
 ---
+# Model Card for PolyLM-Multialpaca
+
+This model is finetuned on [polyLM-13b](https://huggingface.co/DAMO-NLP-MT/polylm-13b) using the following datasets:
+
+
+# Demo
+[Open](https://modelscope.cn/studios/damo/demo-polylm-multialpaca-13b/summary)
+
+# Bias, Risks, and Limitations
+
+The information below in this section are copied from the model's [official model card](https://arxiv.org/pdf/2307.06018.pdf):
+
+> Our contributions are fully methodological: adding the support of multilingualism to LLM during training and SFT phases. It is unavoidable that PolyLM might exhibit several common deficiencies of language models, e.g. hallucination and toxicity. PolyLM should not be used directly in any application, without a prior assessment of safety and fairness concerns specific to the application.
+
+# Citation
+
+**BibTeX:**
+
+```bibtex
+@misc{wei2023polylm,
+    title={PolyLM: An Open Source Polyglot Large Language Model},
+    author={Xiangpeng Wei and Haoran Wei and Huan Lin and Tianhao Li and Pei Zhang and Xingzhang Ren and Mei Li and Yu Wan and Zhiwei Cao and Binbin Xie and Tianxiang Hu and Shangjie Li and Binyuan Hui and Bowen Yu and Dayiheng Liu and Baosong Yang and Fei Huang and Jun Xie},
+    year={2023},
+    eprint={2307.06018},
+    archivePrefix={arXiv},
+    primaryClass={cs.CL}
+}
+```

config.json

@@ -0,0 +1,31 @@
+{
+  "activation_function": "gelu_fast",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.0,
+  "bos_token_id": 255999,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 255999,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_embd": 5120,
+  "n_head": 40,
+  "n_inner": 20480,
+  "n_layer": 40,
+  "n_positions": 2048,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.0,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.0,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "tokenizer_class": "AutoTokenizer",
+  "transformers_version": "4.29.2",
+  "use_cache": true,
+  "vocab_size": 256000
+}

polylm_web_demo_gradio.py

@@ -0,0 +1,171 @@
+import argparse
+import os
+import warnings
+import mdtex2html
+import gradio as gr
+
+import re
+pattern = re.compile("[\n]+")
+
+import torch
+from accelerate import init_empty_weights, load_checkpoint_and_dispatch
+from huggingface_hub import snapshot_download
+
+from transformers.generation.utils import logger
+from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer
+
+
+parser = argparse.ArgumentParser()
+parser.add_argument("--model_name", default=f"DAMO-NLP-MT/polylm-chat-13b",
+                    choices=["DAMO-NLP-MT/polylm-chat-13b"], type=str)
+parser.add_argument("--gpu", default="0", type=str)
+args = parser.parse_args()
+
+os.environ["CUDA_VISIBLE_DEVICES"] = args.gpu
+num_gpus = len(args.gpu.split(","))
+
+if ('int8' in args.model_name or 'int4' in args.model_name) and num_gpus > 1:
+    raise ValueError("Quantized models do not support model parallel. Please run on a single GPU (e.g., --gpu 0).")
+
+logger.setLevel("ERROR")
+warnings.filterwarnings("ignore")
+
+model_path = args.model_name
+if not os.path.exists(args.model_name):
+    model_path = snapshot_download(args.model_name)
+
+config = AutoConfig.from_pretrained(model_path)
+tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=False)
+
+if num_gpus > 1:
+    print("Waiting for all devices to be ready, it may take a few minutes...")
+    with init_empty_weights():
+        raw_model = AutoModelForCausalLM.from_config(config)
+    raw_model.tie_weights()
+    model = load_checkpoint_and_dispatch(
+        raw_model, model_path, device_map="auto", no_split_module_classes=["GPT2Block"]
+    )
+else:
+    print("Loading model files, it may take a few minutes...")
+    model = AutoModelForCausalLM.from_pretrained(model_path, trust_remote_code=True).cuda()
+
+print(model.dtype)
+
+def postprocess(self, y):
+    if y is None:
+        return []
+    for i, (message, response) in enumerate(y):
+        y[i] = (
+            None if message is None else mdtex2html.convert((message)),
+            None if response is None else mdtex2html.convert(response),
+        )
+    return y
+
+
+gr.Chatbot.postprocess = postprocess
+
+
+def parse_text(text):
+    """copy from https://github.com/GaiZhenbiao/ChuanhuChatGPT/"""
+    lines = text.split("\n")
+    lines = [line for line in lines if line != ""]
+    count = 0
+    for i, line in enumerate(lines):
+        if "```" in line:
+            count += 1
+            items = line.split('`')
+            if count % 2 == 1:
+                lines[i] = f'<pre><code class="language-{items[-1]}">'
+            else:
+                lines[i] = f'<br></code></pre>'
+        else:
+            line = line.replace("`", "\`")
+            line = line.replace("<", "&lt;")
+            line = line.replace(">", "&gt;")
+            line = line.replace(" ", "&nbsp;")
+            line = line.replace("*", "&ast;")
+            line = line.replace("_", "&lowbar;")
+            line = line.replace("-", "&#45;")
+            line = line.replace(".", "&#46;")
+            line = line.replace("!", "&#33;")
+            line = line.replace("(", "&#40;")
+            line = line.replace(")", "&#41;")
+            line = line.replace("$", "&#36;")
+            if i > 0:
+                lines[i] = "<br>"+line
+    text = "".join(lines)
+    return text
+
+
+def predict(input, chatbot, max_length, top_p, temperature, history):
+    query = input
+    query = query.strip()
+
+    chatbot.append((query, ""))
+    prompt = ""
+    for i, (old_query, response) in enumerate(history):
+        response = response.strip()
+        prompt += "<|user|>\n" + f"{old_query}\n" + "<|assistant|>\n" + f"{response}\n"
+    prompt += "<|user|>\n" + f"{query}\n" + "<|assistant|>\n"
+    inputs = tokenizer(prompt, return_tensors="pt")
+    with torch.no_grad():
+        outputs = model.generate(
+            inputs.input_ids.cuda(),
+            attention_mask=inputs.attention_mask.cuda(),
+            max_length=max_length,
+            do_sample=True,
+            top_p=top_p,
+            temperature=temperature,
+            repetition_penalty=1.02,
+            num_return_sequences=1,
+            eos_token_id=2,
+            early_stopping=True)
+        response = tokenizer.decode(
+            outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)
+
+    chatbot[-1] = (parse_text(query), parse_text(response.replace("\n ", "\n")))
+    history = history + [(query, response)]
+    print("==========================================================================")
+    print(f"chatbot is {chatbot}")
+    print(f"history is {history}")
+    print("==========================================================================")
+    return chatbot, history
+
+
+def reset_user_input():
+    return gr.update(value='')
+
+
+def reset_state():
+    return [], []
+
+
+with gr.Blocks() as demo:
+    gr.HTML("""<h1 align="center">欢迎使用 PolyLM 多语言人工智能助手！</h1>""")
+
+    chatbot = gr.Chatbot()
+    with gr.Row():
+        with gr.Column(scale=4):
+            with gr.Column(scale=12):
+                user_input = gr.Textbox(show_label=False, placeholder="Input...", lines=10).style(
+                    container=False)
+            with gr.Column(min_width=32, scale=1):
+                submitBtn = gr.Button("Submit", variant="primary")
+        with gr.Column(scale=1):
+            emptyBtn = gr.Button("Clear History")
+            max_length = gr.Slider(
+                0, 4096, value=2048, step=1.0, label="Maximum length", interactive=True)
+            top_p = gr.Slider(0, 1, value=0.8, step=0.01,
+                              label="Top P", interactive=True)
+            temperature = gr.Slider(
+                0, 1, value=0.7, step=0.01, label="Temperature", interactive=True)
+
+    history = gr.State([])  # (message, bot_message)
+
+    submitBtn.click(predict, [user_input, chatbot, max_length, top_p, temperature, history], [chatbot, history],
+                    show_progress=True)
+    submitBtn.click(reset_user_input, [], [user_input])
+
+    emptyBtn.click(reset_state, outputs=[chatbot, history], show_progress=True)
+
+demo.queue().launch(share=False, inbrowser=True)

pytorch_model-00001-of-00004.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f783ac6c62c466107986de443a7e543dc18a70ab848cf8a9bbb897a39c8cd97
+size 8809402977

pytorch_model-00002-of-00004.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af6e958d9f17bf1b71c6b434ba0de98d43ca76cfaefe143a9c1377bd08c23323
+size 8927403605

pytorch_model-00003-of-00004.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc793bbf117cafdb183affe11ba3927cae91d31fd5e776ef9c7fe635115cbda3
+size 8927403605

pytorch_model-00004-of-00004.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d08f10b01ecd6dce32c434f7148dce0873445bddf3e00082fa85ce93d80d377
+size 4106529922

pytorch_model.bin.index.json

@@ -0,0 +1,572 @@
+{
+  "metadata": {
+    "total_size": 30770565200
+  },
+  "weight_map": {
+    "lm_head.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.0.attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.attn.c_attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.attn.c_attn.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.attn.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.attn.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.attn.masked_bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.ln_1.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.ln_1.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.ln_2.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.ln_2.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.mlp.c_fc.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.mlp.c_fc.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.mlp.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.0.mlp.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.attn.c_attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.attn.c_attn.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.attn.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.attn.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.attn.masked_bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.ln_1.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.ln_1.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.ln_2.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.ln_2.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.mlp.c_fc.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.mlp.c_fc.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.mlp.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.1.mlp.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.10.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.10.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.11.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.12.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.13.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.14.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.15.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.16.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.17.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.18.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.19.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.2.attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.attn.c_attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.attn.c_attn.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.attn.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.attn.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.attn.masked_bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.ln_1.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.ln_1.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.ln_2.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.ln_2.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.mlp.c_fc.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.mlp.c_fc.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.mlp.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.2.mlp.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.20.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.20.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.21.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.22.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.attn.c_attn.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.attn.c_attn.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.attn.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.attn.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.attn.masked_bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.ln_1.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.ln_1.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.ln_2.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.ln_2.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.mlp.c_fc.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.mlp.c_fc.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.23.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.23.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.24.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.25.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.26.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.27.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.28.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.29.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.3.attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.attn.c_attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.attn.c_attn.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.attn.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.attn.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.attn.masked_bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.ln_1.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.ln_1.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.ln_2.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.ln_2.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.mlp.c_fc.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.mlp.c_fc.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.mlp.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.3.mlp.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.30.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.30.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.31.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.32.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.33.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.34.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.35.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.mlp.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.36.mlp.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.attn.c_attn.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.attn.c_attn.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.attn.c_proj.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.attn.c_proj.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.attn.masked_bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.ln_1.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.ln_1.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.ln_2.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.ln_2.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.mlp.c_fc.bias": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.mlp.c_fc.weight": "pytorch_model-00003-of-00004.bin",
+    "transformer.h.37.mlp.c_proj.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.37.mlp.c_proj.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.attn.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.attn.c_attn.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.attn.c_attn.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.attn.c_proj.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.attn.c_proj.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.attn.masked_bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.ln_1.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.ln_1.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.ln_2.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.ln_2.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.mlp.c_fc.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.mlp.c_fc.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.mlp.c_proj.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.38.mlp.c_proj.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.attn.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.attn.c_attn.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.attn.c_attn.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.attn.c_proj.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.attn.c_proj.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.attn.masked_bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.ln_1.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.ln_1.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.ln_2.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.ln_2.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.mlp.c_fc.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.mlp.c_fc.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.mlp.c_proj.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.39.mlp.c_proj.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.h.4.attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.attn.c_attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.attn.c_attn.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.attn.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.attn.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.attn.masked_bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.ln_1.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.ln_1.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.ln_2.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.ln_2.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.mlp.c_fc.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.mlp.c_fc.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.mlp.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.4.mlp.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.attn.c_attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.attn.c_attn.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.attn.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.attn.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.attn.masked_bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.ln_1.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.ln_1.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.ln_2.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.ln_2.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.mlp.c_fc.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.mlp.c_fc.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.mlp.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.5.mlp.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.attn.c_attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.attn.c_attn.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.attn.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.attn.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.attn.masked_bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.ln_1.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.ln_1.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.ln_2.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.ln_2.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.mlp.c_fc.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.mlp.c_fc.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.mlp.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.6.mlp.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.attn.c_attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.attn.c_attn.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.attn.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.attn.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.attn.masked_bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.ln_1.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.ln_1.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.ln_2.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.ln_2.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.mlp.c_fc.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.mlp.c_fc.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.mlp.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.7.mlp.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.attn.c_attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.attn.c_attn.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.attn.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.attn.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.attn.masked_bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.ln_1.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.ln_1.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.ln_2.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.ln_2.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.mlp.c_fc.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.mlp.c_fc.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.mlp.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.8.mlp.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.attn.c_attn.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.attn.c_attn.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.attn.c_proj.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.attn.c_proj.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.attn.masked_bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.ln_1.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.ln_1.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.ln_2.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.ln_2.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.mlp.c_fc.bias": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.mlp.c_fc.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.h.9.mlp.c_proj.bias": "pytorch_model-00002-of-00004.bin",
+    "transformer.h.9.mlp.c_proj.weight": "pytorch_model-00002-of-00004.bin",
+    "transformer.ln_f.bias": "pytorch_model-00004-of-00004.bin",
+    "transformer.ln_f.weight": "pytorch_model-00004-of-00004.bin",
+    "transformer.wpe.weight": "pytorch_model-00001-of-00004.bin",
+    "transformer.wte.weight": "pytorch_model-00001-of-00004.bin"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "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
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63639ba62c4af921cb7c373630bee7d45568fc5ff7a98add756e8aa57df59065
+size 4754121

tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "add_bos_token": false,
+  "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
+  },
+  "model_max_length": 2048,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": {
+    "__type": "AddedToken",
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}