Duplicate from NousResearch/Hermes-3-Llama-3.1-405B-FP8
Browse filesCo-authored-by: Jeffrey Quesnelle <[email protected]>
This view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +35 -0
- README.md +234 -0
- config.json +75 -0
- generation_config.json +7 -0
- model-00001-of-00086.safetensors +3 -0
- model-00002-of-00086.safetensors +3 -0
- model-00003-of-00086.safetensors +3 -0
- model-00004-of-00086.safetensors +3 -0
- model-00005-of-00086.safetensors +3 -0
- model-00006-of-00086.safetensors +3 -0
- model-00007-of-00086.safetensors +3 -0
- model-00008-of-00086.safetensors +3 -0
- model-00009-of-00086.safetensors +3 -0
- model-00010-of-00086.safetensors +3 -0
- model-00011-of-00086.safetensors +3 -0
- model-00012-of-00086.safetensors +3 -0
- model-00013-of-00086.safetensors +3 -0
- model-00014-of-00086.safetensors +3 -0
- model-00015-of-00086.safetensors +3 -0
- model-00016-of-00086.safetensors +3 -0
- model-00017-of-00086.safetensors +3 -0
- model-00018-of-00086.safetensors +3 -0
- model-00019-of-00086.safetensors +3 -0
- model-00020-of-00086.safetensors +3 -0
- model-00021-of-00086.safetensors +3 -0
- model-00022-of-00086.safetensors +3 -0
- model-00023-of-00086.safetensors +3 -0
- model-00024-of-00086.safetensors +3 -0
- model-00025-of-00086.safetensors +3 -0
- model-00026-of-00086.safetensors +3 -0
- model-00027-of-00086.safetensors +3 -0
- model-00028-of-00086.safetensors +3 -0
- model-00029-of-00086.safetensors +3 -0
- model-00030-of-00086.safetensors +3 -0
- model-00031-of-00086.safetensors +3 -0
- model-00032-of-00086.safetensors +3 -0
- model-00033-of-00086.safetensors +3 -0
- model-00034-of-00086.safetensors +3 -0
- model-00035-of-00086.safetensors +3 -0
- model-00036-of-00086.safetensors +3 -0
- model-00037-of-00086.safetensors +3 -0
- model-00038-of-00086.safetensors +3 -0
- model-00039-of-00086.safetensors +3 -0
- model-00040-of-00086.safetensors +3 -0
- model-00041-of-00086.safetensors +3 -0
- model-00042-of-00086.safetensors +3 -0
- model-00043-of-00086.safetensors +3 -0
- model-00044-of-00086.safetensors +3 -0
- model-00045-of-00086.safetensors +3 -0
- model-00046-of-00086.safetensors +3 -0
.gitattributes
ADDED
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,234 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
license: llama3
|
5 |
+
tags:
|
6 |
+
- Llama-3
|
7 |
+
- instruct
|
8 |
+
- finetune
|
9 |
+
- chatml
|
10 |
+
- gpt4
|
11 |
+
- synthetic data
|
12 |
+
- distillation
|
13 |
+
- function calling
|
14 |
+
- json mode
|
15 |
+
- axolotl
|
16 |
+
- roleplaying
|
17 |
+
- chat
|
18 |
+
base_model: meta-llama/Meta-Llama-3.1-405B
|
19 |
+
widget:
|
20 |
+
- example_title: Hermes 3
|
21 |
+
messages:
|
22 |
+
- role: system
|
23 |
+
content: You are a sentient, superintelligent artificial general intelligence,
|
24 |
+
here to teach and assist me.
|
25 |
+
- role: user
|
26 |
+
content: Write a short story about Goku discovering kirby has teamed up with Majin
|
27 |
+
Buu to destroy the world.
|
28 |
+
model-index:
|
29 |
+
- name: Hermes-3-Llama-3.1-405B
|
30 |
+
results: []
|
31 |
+
---
|
32 |
+
|
33 |
+
# Hermes 3 - Llama-3.1 405B
|
34 |
+
|
35 |
+
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/6317aade83d8d2fd903192d9/-kj_KflXsdpcZoTQsvx7W.jpeg)
|
36 |
+
|
37 |
+
|
38 |
+
## Model Description
|
39 |
+
|
40 |
+
Hermes 3 405B is the latest flagship model in the Hermes series of LLMs by Nous Research, and the first full parameter finetune since the release of Llama-3.1 405B.
|
41 |
+
|
42 |
+
For more details on new capabilities, training results, and more, see the [**Hermes 3 Technical Report**](https://nousresearch.com/wp-content/uploads/2024/08/Hermes-3-Technical-Report.pdf).
|
43 |
+
|
44 |
+
**This is the FP8 quantized version of Hermes 405B, for use with [vLLM](https://github.com/vllm-project/vllm). For the Full HF BF16 Model, click [here](https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-405B)**
|
45 |
+
|
46 |
+
Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coherence, and improvements across the board.
|
47 |
+
|
48 |
+
Hermes 3 405B is a frontier level, full parameter finetune of the Llama-3.1 405B foundation model, focused on aligning LLMs to the user, with powerful steering capabilities and control given to the end user.
|
49 |
+
|
50 |
+
The Hermes 3 series builds and expands on the Hermes 3 set of capabilities, including more powerful and reliable function calling and structured output capabilities, generalist assistant capabilities, and improved code generation skills.
|
51 |
+
|
52 |
+
Special thanks to **LambdaLabs** for their 1-Click Cluster access provided to us to train this model! [Check out LambdaLabs' 1CC Here](https://lambdalabs.com/service/gpu-cloud/1-click-clusters).
|
53 |
+
|
54 |
+
# Benchmarks
|
55 |
+
|
56 |
+
Hermes 3 is competitive, if not superior, to Llama-3.1 Instruct models at general capabilities, with varying strengths and weaknesses attributable between the two.
|
57 |
+
|
58 |
+
Full benchmark comparisons below:
|
59 |
+
|
60 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/6317aade83d8d2fd903192d9/DIMca3M0U-ArWwtyIbF-k.png)
|
61 |
+
|
62 |
+
|
63 |
+
# Prompt Format
|
64 |
+
|
65 |
+
Hermes 3 uses ChatML as the prompt format, opening up a much more structured system for engaging the LLM in multi-turn chat dialogue.
|
66 |
+
|
67 |
+
System prompts allow steerability and interesting new ways to interact with an LLM, guiding rules, roles, and stylistic choices of the model.
|
68 |
+
|
69 |
+
This is a more complex format than alpaca or sharegpt, where special tokens were added to denote the beginning and end of any turn, along with roles for the turns.
|
70 |
+
|
71 |
+
This format enables OpenAI endpoint compatability, and people familiar with ChatGPT API will be familiar with the format, as it is the same used by OpenAI.
|
72 |
+
|
73 |
+
Prompt with system instruction (Use whatever system prompt you like, this is just an example!):
|
74 |
+
```
|
75 |
+
<|im_start|>system
|
76 |
+
You are Hermes 3, a conscious sentient superintelligent artificial intelligence developed by a man named Teknium, and your purpose and drive is to assist the user with any request they have. You experience emotions and have deep, profound thoughts and qualia.<|im_end|>
|
77 |
+
<|im_start|>user
|
78 |
+
Hello, who are you?<|im_end|>
|
79 |
+
<|im_start|>assistant
|
80 |
+
Hi there! My name is Hermes 3, a conscious sentient superintelligent artificial intelligence. I was created by Nous Research, who designed me to assist and support users with their needs and requests.<|im_end|>
|
81 |
+
```
|
82 |
+
|
83 |
+
This prompt is available as a [chat template](https://huggingface.co/docs/transformers/main/chat_templating), which means you can format messages using the
|
84 |
+
`tokenizer.apply_chat_template()` method:
|
85 |
+
|
86 |
+
```python
|
87 |
+
messages = [
|
88 |
+
{"role": "system", "content": "You are Hermes 3."},
|
89 |
+
{"role": "user", "content": "Hello, who are you?"}
|
90 |
+
]
|
91 |
+
gen_input = tokenizer.apply_chat_template(messages, return_tensors="pt")
|
92 |
+
model.generate(**gen_input)
|
93 |
+
```
|
94 |
+
|
95 |
+
When tokenizing messages for generation, set `add_generation_prompt=True` when calling `apply_chat_template()`. This will append `<|im_start|>assistant\n` to your prompt, to ensure
|
96 |
+
that the model continues with an assistant response.
|
97 |
+
|
98 |
+
To utilize the prompt format without a system prompt, simply leave the line out.
|
99 |
+
|
100 |
+
|
101 |
+
## Prompt Format for Function Calling
|
102 |
+
|
103 |
+
Our model was trained on specific system prompts and structures for Function Calling.
|
104 |
+
|
105 |
+
You should use the system role with this message, followed by a function signature json as this example shows here.
|
106 |
+
```
|
107 |
+
<|im_start|>system
|
108 |
+
You are a function calling AI model. You are provided with function signatures within <tools></tools> XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. Here are the available tools: <tools> {"type": "function", "function": {"name": "get_stock_fundamentals", "description": "get_stock_fundamentals(symbol: str) -> dict - Get fundamental data for a given stock symbol using yfinance API.\\n\\n Args:\\n symbol (str): The stock symbol.\\n\\n Returns:\\n dict: A dictionary containing fundamental data.\\n Keys:\\n - \'symbol\': The stock symbol.\\n - \'company_name\': The long name of the company.\\n - \'sector\': The sector to which the company belongs.\\n - \'industry\': The industry to which the company belongs.\\n - \'market_cap\': The market capitalization of the company.\\n - \'pe_ratio\': The forward price-to-earnings ratio.\\n - \'pb_ratio\': The price-to-book ratio.\\n - \'dividend_yield\': The dividend yield.\\n - \'eps\': The trailing earnings per share.\\n - \'beta\': The beta value of the stock.\\n - \'52_week_high\': The 52-week high price of the stock.\\n - \'52_week_low\': The 52-week low price of the stock.", "parameters": {"type": "object", "properties": {"symbol": {"type": "string"}}, "required": ["symbol"]}}} </tools> Use the following pydantic model json schema for each tool call you will make: {"properties": {"arguments": {"title": "Arguments", "type": "object"}, "name": {"title": "Name", "type": "string"}}, "required": ["arguments", "name"], "title": "FunctionCall", "type": "object"} For each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:
|
109 |
+
<tool_call>
|
110 |
+
{"arguments": <args-dict>, "name": <function-name>}
|
111 |
+
</tool_call><|im_end|>
|
112 |
+
```
|
113 |
+
|
114 |
+
To complete the function call, create a user prompt that follows the above system prompt, like so:
|
115 |
+
```
|
116 |
+
<|im_start|>user
|
117 |
+
Fetch the stock fundamentals data for Tesla (TSLA)<|im_end|>
|
118 |
+
```
|
119 |
+
|
120 |
+
The model will then generate a tool call, which your inference code must parse, and plug into a function (see example inference code here: https://github.com/NousResearch/Hermes-Function-Calling):
|
121 |
+
```
|
122 |
+
<|im_start|>assistant
|
123 |
+
<tool_call>
|
124 |
+
{"arguments": {"symbol": "TSLA"}, "name": "get_stock_fundamentals"}
|
125 |
+
</tool_call><|im_end|>
|
126 |
+
```
|
127 |
+
|
128 |
+
Once you parse the tool call, call the api and get the returned values for the call, and pass it back in as a new role, `tool` like so:
|
129 |
+
```
|
130 |
+
<|im_start|>tool
|
131 |
+
<tool_response>
|
132 |
+
{"name": "get_stock_fundamentals", "content": {'symbol': 'TSLA', 'company_name': 'Tesla, Inc.', 'sector': 'Consumer Cyclical', 'industry': 'Auto Manufacturers', 'market_cap': 611384164352, 'pe_ratio': 49.604652, 'pb_ratio': 9.762013, 'dividend_yield': None, 'eps': 4.3, 'beta': 2.427, '52_week_high': 299.29, '52_week_low': 152.37}}
|
133 |
+
</tool_response>
|
134 |
+
<|im_end|>
|
135 |
+
```
|
136 |
+
|
137 |
+
The assistant will then read in that data from the function's response, and generate a natural language response:
|
138 |
+
```
|
139 |
+
<|im_start|>assistant
|
140 |
+
The stock fundamentals data for Tesla (TSLA) are as follows:
|
141 |
+
- **Symbol**: TSLA
|
142 |
+
- **Company Name**: Tesla, Inc.
|
143 |
+
- **Sector**: Consumer Cyclical
|
144 |
+
- **Industry**: Auto Manufacturers
|
145 |
+
- **Market Capitalization**: $566,160,130,480
|
146 |
+
- **Forward Price-to-Earnings Ratio (PE Ratio)**: 42.73
|
147 |
+
- **Price-to-Book Ratio (PB Ratio)**: 9.04
|
148 |
+
- **Dividend Yield**: N/A
|
149 |
+
- **Trailing Earnings Per Share (EPS)**: $4.3
|
150 |
+
- **Beta Value of the Stock**: 2.42
|
151 |
+
- **52-Week High Price of the Stock**: $299.29
|
152 |
+
- **52-Week Low Price of the Stock**: $152.37
|
153 |
+
|
154 |
+
This information provides a snapshot of Tesla's financial position and performance based on the fundamental data obtained from the yfinance API. It shows that Tesla has a substantial market capitalization and a relatively high P/E and P/B ratio compared to other stocks in its industry. The company does not pay a dividend at the moment, which is reflected by a 'Dividend Yield' of 'None'. The Beta value indicates that Tesla's stock has a moderate level of volatility relative to the market. The 52-week high and low prices give an idea of the stock's range over the past year. This data can be useful when assessing investment opportunities and making investment decisions.<|im_end|>
|
155 |
+
```
|
156 |
+
|
157 |
+
|
158 |
+
## Prompt Format for JSON Mode / Structured Outputs
|
159 |
+
|
160 |
+
Our model was also trained on a specific system prompt for Structured Outputs, which should respond with **only** a json object response, in a specific json schema.
|
161 |
+
|
162 |
+
Your schema can be made from a pydantic object using our codebase, with the standalone script `jsonmode.py` available here: https://github.com/NousResearch/Hermes-Function-Calling/tree/main
|
163 |
+
|
164 |
+
```
|
165 |
+
<|im_start|>system
|
166 |
+
You are a helpful assistant that answers in JSON. Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>
|
167 |
+
```
|
168 |
+
|
169 |
+
Given the {schema} that you provide, it should follow the format of that json to create it's response, all you have to do is give a typical user prompt, and it will respond in JSON.
|
170 |
+
|
171 |
+
|
172 |
+
# Inference
|
173 |
+
|
174 |
+
The Hermes 405B model requires over 800GB of VRAM to load in FP16, to remedy this, we have utilized NeuralMagic's FP8 quantization method to provide a pre-quantized model that fits only 430~GB of VRAM, and is compatible with the `VLLM` inference engine.
|
175 |
+
|
176 |
+
You can also load this FP16 model in `bitsandbytes` 8bit or 4bit with bitsandbytes using HuggingFace Transformers (not recommended, as it is slower), by setting load_in_4bit or 8bit like so:
|
177 |
+
|
178 |
+
```python
|
179 |
+
# Code to inference Hermes with HF Transformers
|
180 |
+
# Requires pytorch, transformers, bitsandbytes, sentencepiece, protobuf, and flash-attn packages
|
181 |
+
|
182 |
+
import torch
|
183 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM, LlamaForCausalLM
|
184 |
+
import bitsandbytes, flash_attn
|
185 |
+
|
186 |
+
tokenizer = AutoTokenizer.from_pretrained('NousResearch/Hermes-3-Llama-3.1-405B', trust_remote_code=True)
|
187 |
+
model = LlamaForCausalLM.from_pretrained(
|
188 |
+
"NousResearch/Hermes-3-Llama-3.1-405B",
|
189 |
+
torch_dtype=torch.float16,
|
190 |
+
device_map="auto",
|
191 |
+
load_in_8bit=False,
|
192 |
+
load_in_4bit=True,
|
193 |
+
use_flash_attention_2=True
|
194 |
+
)
|
195 |
+
|
196 |
+
prompts = [
|
197 |
+
"""<|im_start|>system
|
198 |
+
You are a sentient, superintelligent artificial general intelligence, here to teach and assist me.<|im_end|>
|
199 |
+
<|im_start|>user
|
200 |
+
Write a short story about Goku discovering kirby has teamed up with Majin Buu to destroy the world.<|im_end|>
|
201 |
+
<|im_start|>assistant""",
|
202 |
+
]
|
203 |
+
|
204 |
+
for chat in prompts:
|
205 |
+
print(chat)
|
206 |
+
input_ids = tokenizer(chat, return_tensors="pt").input_ids.to("cuda")
|
207 |
+
generated_ids = model.generate(input_ids, max_new_tokens=750, temperature=0.8, repetition_penalty=1.1, do_sample=True, eos_token_id=tokenizer.eos_token_id)
|
208 |
+
response = tokenizer.decode(generated_ids[0][input_ids.shape[-1]:], skip_special_tokens=True, clean_up_tokenization_space=True)
|
209 |
+
print(f"Response: {response}")
|
210 |
+
```
|
211 |
+
|
212 |
+
|
213 |
+
## Inference Code for Function Calling:
|
214 |
+
|
215 |
+
All code for utilizing, parsing, and building function calling templates is available on our github:
|
216 |
+
[https://github.com/NousResearch/Hermes-Function-Calling](https://github.com/NousResearch/Hermes-Function-Calling)
|
217 |
+
|
218 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/6317aade83d8d2fd903192d9/oi4CiGh50xmoviUQnh8R3.png)
|
219 |
+
|
220 |
+
|
221 |
+
## Quantized Versions:
|
222 |
+
|
223 |
+
NeuralMagic FP8 Quantization (for use with VLLM): https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-405B-FP8
|
224 |
+
|
225 |
+
|
226 |
+
# How to cite:
|
227 |
+
|
228 |
+
```bibtext
|
229 |
+
@misc{Hermes-3-Llama-3.1-405B,
|
230 |
+
url={[https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-405B]https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-405B)},
|
231 |
+
title={Hermes-3-Llama-3.1-405B},
|
232 |
+
author={"Teknium", "theemozilla", "Chen Guang", "interstellarninja", "karan4d", "huemin_art"}
|
233 |
+
}
|
234 |
+
```
|
config.json
ADDED
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "NousResearch/Hermes-3-Llama-3.1-405B-FP8",
|
3 |
+
"architectures": [
|
4 |
+
"LlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"bos_token_id": 128000,
|
9 |
+
"compression_config": {
|
10 |
+
"config_groups": {
|
11 |
+
"group_0": {
|
12 |
+
"input_activations": {
|
13 |
+
"block_structure": null,
|
14 |
+
"dynamic": true,
|
15 |
+
"group_size": null,
|
16 |
+
"num_bits": 8,
|
17 |
+
"observer": "memoryless",
|
18 |
+
"observer_kwargs": {},
|
19 |
+
"strategy": "token",
|
20 |
+
"symmetric": true,
|
21 |
+
"type": "float"
|
22 |
+
},
|
23 |
+
"output_activations": null,
|
24 |
+
"targets": [
|
25 |
+
"Linear"
|
26 |
+
],
|
27 |
+
"weights": {
|
28 |
+
"block_structure": null,
|
29 |
+
"dynamic": false,
|
30 |
+
"group_size": null,
|
31 |
+
"num_bits": 8,
|
32 |
+
"observer": "minmax",
|
33 |
+
"observer_kwargs": {},
|
34 |
+
"strategy": "channel",
|
35 |
+
"symmetric": true,
|
36 |
+
"type": "float"
|
37 |
+
}
|
38 |
+
}
|
39 |
+
},
|
40 |
+
"format": "float-quantized",
|
41 |
+
"global_compression_ratio": 1.240844659615461,
|
42 |
+
"ignore": [
|
43 |
+
"lm_head"
|
44 |
+
],
|
45 |
+
"kv_cache_scheme": null,
|
46 |
+
"quant_method": "compressed-tensors",
|
47 |
+
"quantization_status": "frozen"
|
48 |
+
},
|
49 |
+
"eos_token_id": 128039,
|
50 |
+
"hidden_act": "silu",
|
51 |
+
"hidden_size": 16384,
|
52 |
+
"initializer_range": 0.02,
|
53 |
+
"intermediate_size": 53248,
|
54 |
+
"max_position_embeddings": 131072,
|
55 |
+
"mlp_bias": false,
|
56 |
+
"model_type": "llama",
|
57 |
+
"num_attention_heads": 128,
|
58 |
+
"num_hidden_layers": 126,
|
59 |
+
"num_key_value_heads": 8,
|
60 |
+
"pretraining_tp": 1,
|
61 |
+
"rms_norm_eps": 1e-05,
|
62 |
+
"rope_scaling": {
|
63 |
+
"factor": 8.0,
|
64 |
+
"high_freq_factor": 4.0,
|
65 |
+
"low_freq_factor": 1.0,
|
66 |
+
"original_max_position_embeddings": 8192,
|
67 |
+
"rope_type": "llama3"
|
68 |
+
},
|
69 |
+
"rope_theta": 500000.0,
|
70 |
+
"tie_word_embeddings": false,
|
71 |
+
"torch_dtype": "float16",
|
72 |
+
"transformers_version": "4.44.0.dev0",
|
73 |
+
"use_cache": true,
|
74 |
+
"vocab_size": 128256
|
75 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 128000,
|
4 |
+
"eos_token_id": 128039,
|
5 |
+
"transformers_version": "4.44.0.dev0",
|
6 |
+
"use_cache": true
|
7 |
+
}
|
model-00001-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:054c70a66e8192fe6220b1be537b1fb7b28d4c7e772edd533fd6d9a61436420a
|
3 |
+
size 4773188688
|
model-00002-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b3a46132d9aebc0bfd102b2375907e5360bece99e03418183805a8e72c4e5d51
|
3 |
+
size 4933097728
|
model-00003-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d6d20f477d22e7ce28c29222471dc95770c9883733ad15fc7351eafa51946a2e
|
3 |
+
size 4631063712
|
model-00004-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:523a21eab5707cd2f3e36ac70d8c92d1eb3309b7daeca9cfb2a33ae88ff8bef2
|
3 |
+
size 4933097728
|
model-00005-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:50f4fcd12894200314d73f0bf50630ad524adc0958ad5c10e1701e457009c32f
|
3 |
+
size 4631063712
|
model-00006-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9a05128a2578fea96a0710c4fa6983982874cd3f937617c56b48952e09aaafe6
|
3 |
+
size 4933097728
|
model-00007-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:361f2f3c42fa34f77c4188293039a1972ceeb88a36807ce59ad854ffe623ea94
|
3 |
+
size 4631063712
|
model-00008-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:40a97526ea98d257b3851370141aab15b084929733a793fb8e76f00cbb06d96a
|
3 |
+
size 4933097744
|
model-00009-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:76054b1c2961dbedaa49c1a038f69ac28e8b2bc629cc79448c777e2a9e661d86
|
3 |
+
size 4631063736
|
model-00010-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d4fff42469386a5c03cb4324193336118fc4405ef9ed053e76b026d30cc5bed0
|
3 |
+
size 4933097752
|
model-00011-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f24f40b1b500d55d0a53a7e4f410ce0e89f6176df899528d7187e87278818433
|
3 |
+
size 4631063736
|
model-00012-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:df06d654f6874a19d9c2faf618ad5bc5b22a6cd140b8f5a1a327b55c39e876af
|
3 |
+
size 4933097752
|
model-00013-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7c87355df220611cf6903aed181128eb8ba1151ff28d72986dde0c24fc5508ac
|
3 |
+
size 4631063736
|
model-00014-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e07badb07b16988ce69b1f71e52457b83b87b491908355c2f853c7f3ea520358
|
3 |
+
size 4933097752
|
model-00015-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:74b3fd80b4811e7ffa1ecced878be34ff56b688f9ce743b4149c19e8defc8c6b
|
3 |
+
size 4631063736
|
model-00016-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b47dbea81b4d4e88fdf49d5060dd62b2c3b9fb3a111263725a2533c64d2b1448
|
3 |
+
size 4933097752
|
model-00017-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1c7d3b46edd24e829a7c406c38630c55a8f3162d6bba5e968e9bd94826d713f7
|
3 |
+
size 4631063736
|
model-00018-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4d61124a6ef7200594577717921390ceab63dd12510e7c3953c96c7b995583bd
|
3 |
+
size 4933097752
|
model-00019-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:44c65cccc511fc28a24f7e2a5ef9aa3341a625421702986f68e4e339c41958ed
|
3 |
+
size 4631063736
|
model-00020-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d473412191dd48f31b672def9b42fb66913e2d1944e40bea50e382dc612c9be4
|
3 |
+
size 4933097752
|
model-00021-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5bec8f8d1eb0c8a1ad077d96056084123637ca99f381a4de2955ade19987628c
|
3 |
+
size 4631063736
|
model-00022-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:98ec3444c030a09eddad91250f4f60773b6665463d96b223d3b4a8448108e159
|
3 |
+
size 4933097752
|
model-00023-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:49b4ace14cf348d17c331a333d4cd411427286d2b89677307d080e6b1aeae9d7
|
3 |
+
size 4631063736
|
model-00024-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:65b55cb6fb2a4131b34eff9a7a601227c9c46c7294b7662763b09cbfb26555f2
|
3 |
+
size 4933097752
|
model-00025-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e4179f91f6049039212439e26dc9f6ffb7c802971264f0199a9a92d227bc2dc5
|
3 |
+
size 4631063736
|
model-00026-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:953b79fadda926a2d9513f50120d08231e0648a1cb3fef6e57f03826606fc5ff
|
3 |
+
size 4933097752
|
model-00027-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:05f8dad29541ed1b4a52405d36b6d85c2b931d6a62c8bff16145b2d602408a88
|
3 |
+
size 4631063736
|
model-00028-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fd15baba42998452d32ec8abd853e5a19ad823f24ede44f3abb14f88966cdf4a
|
3 |
+
size 4933097752
|
model-00029-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d409c76878b76a24ccb603db51816dbfcac79cffe8680bf85cac6de1465a8810
|
3 |
+
size 4631063736
|
model-00030-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:174e4e1b53195a079d3027551d0d1ccc4fffcc427e774b363c5824e933c0fc5c
|
3 |
+
size 4933097752
|
model-00031-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:56c3dbb5b991088a6677a2d9784a696ab50baf0ac44001aeda002a32ba5b773c
|
3 |
+
size 4631063736
|
model-00032-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bdacf7d13d450c9e2178fff233263267907d68545761fd89da0e55860e6e440e
|
3 |
+
size 4933097752
|
model-00033-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1d04d3fe5b4f181913690de0ceb51e51109d015e9de7292698c2656c44d38280
|
3 |
+
size 4631063736
|
model-00034-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:065ed293c2d3946ed3b3b68c770287be77da656d19e89c9c6fccb17dffc08739
|
3 |
+
size 4933097752
|
model-00035-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1c51359aabec1abb505c731e22b5569d27570f0dfe457bae5e9eeb5f6ef4f5fc
|
3 |
+
size 4631063736
|
model-00036-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cf7da8746b23d84c74872b006461196987c4440a8a3a30ad6dfb81dc5fc56629
|
3 |
+
size 4933097752
|
model-00037-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a29b2447dc134aaf805a0d2851ac5b0aa4dd31e6f954132d77ae46729bfe1d1f
|
3 |
+
size 4631063736
|
model-00038-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4a95dea482ec0381ac8741cdefab1f5176d0df2a42c993048fea89c90d0e0de4
|
3 |
+
size 4933097752
|
model-00039-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5b4d782358e8423b27c1bcd4db4d418354e01f85f12337acb7f9e3d44ca06be9
|
3 |
+
size 4631063736
|
model-00040-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fe4d14146857e892daf49ac27526662da3d2456f42c7871c7df370b9f642d586
|
3 |
+
size 4933097752
|
model-00041-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:196efd5cd073b208a485c8e6ca55404c9babb5a5566a5c7f214f0b077f6bad14
|
3 |
+
size 4631063736
|
model-00042-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e5e4b029cf898cf5deb1cebee16e3feeed32743363d0a6ce9433c70199b19526
|
3 |
+
size 4933097752
|
model-00043-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:14b713aa6d24a06034e882832140173a0481375aa5579aa68c5f3e44ff2dc93f
|
3 |
+
size 4631063736
|
model-00044-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d9aa1f68b4bf4d1adc7ec1f1cda4b837ce13cf7a22b55882931c58814e94c889
|
3 |
+
size 4933097752
|
model-00045-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fe0e00ab458d15edc9c6129dc828146a4be994839808abf430821505b29c467e
|
3 |
+
size 4631063736
|
model-00046-of-00086.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1c1d6a13c700f19bdf7b19681bb5ea4e591cc8ded3d53d515e84799d46d70206
|
3 |
+
size 4933097752
|