aashish1904
commited on
Commit
•
bb39ff0
1
Parent(s):
49013fc
Upload README.md with huggingface_hub
Browse files
README.md
ADDED
@@ -0,0 +1,239 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
|
2 |
+
---
|
3 |
+
|
4 |
+
language:
|
5 |
+
- en
|
6 |
+
license: llama3
|
7 |
+
tags:
|
8 |
+
- Llama-3
|
9 |
+
- instruct
|
10 |
+
- finetune
|
11 |
+
- chatml
|
12 |
+
- gpt4
|
13 |
+
- synthetic data
|
14 |
+
- distillation
|
15 |
+
- function calling
|
16 |
+
- json mode
|
17 |
+
- axolotl
|
18 |
+
- roleplaying
|
19 |
+
- chat
|
20 |
+
base_model: meta-llama/Meta-Llama-3.1-8B
|
21 |
+
widget:
|
22 |
+
- example_title: Hermes 3
|
23 |
+
messages:
|
24 |
+
- role: system
|
25 |
+
content: You are a sentient, superintelligent artificial general intelligence,
|
26 |
+
here to teach and assist me.
|
27 |
+
- role: user
|
28 |
+
content: What is the meaning of life?
|
29 |
+
model-index:
|
30 |
+
- name: Hermes-3-Llama-3.1-70B
|
31 |
+
results: []
|
32 |
+
|
33 |
+
---
|
34 |
+
|
35 |
+
![](https://lh7-rt.googleusercontent.com/docsz/AD_4nXeiuCm7c8lEwEJuRey9kiVZsRn2W-b4pWlu3-X534V3YmVuVc2ZL-NXg2RkzSOOS2JXGHutDuyyNAUtdJI65jGTo8jT9Y99tMi4H4MqL44Uc5QKG77B0d6-JfIkZHFaUA71-RtjyYZWVIhqsNZcx8-OMaA?key=xt3VSDoCbmTY7o-cwwOFwQ)
|
36 |
+
|
37 |
+
# QuantFactory/Hermes-3-Llama-3.1-8B-GGUF
|
38 |
+
This is quantized version of [NousResearch/Hermes-3-Llama-3.1-8B](https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B) created using llama.cpp
|
39 |
+
|
40 |
+
# Original Model Card
|
41 |
+
|
42 |
+
# Hermes 3 - Llama-3.1 8B
|
43 |
+
|
44 |
+
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/6317aade83d8d2fd903192d9/bMcZ3sNNQK8SRZpHXBmwM.jpeg)
|
45 |
+
|
46 |
+
## Model Description
|
47 |
+
|
48 |
+
Hermes 3 is the latest version of our flagship Hermes series of LLMs by Nous Research.
|
49 |
+
|
50 |
+
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).
|
51 |
+
|
52 |
+
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.
|
53 |
+
|
54 |
+
The ethos of the Hermes series of models is focused on aligning LLMs to the user, with powerful steering capabilities and control given to the end user.
|
55 |
+
|
56 |
+
The Hermes 3 series builds and expands on the Hermes 2 set of capabilities, including more powerful and reliable function calling and structured output capabilities, generalist assistant capabilities, and improved code generation skills.
|
57 |
+
|
58 |
+
|
59 |
+
# Benchmarks
|
60 |
+
|
61 |
+
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.
|
62 |
+
|
63 |
+
Full benchmark comparisons below:
|
64 |
+
|
65 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/6317aade83d8d2fd903192d9/DIMca3M0U-ArWwtyIbF-k.png)
|
66 |
+
|
67 |
+
|
68 |
+
# Prompt Format
|
69 |
+
|
70 |
+
Hermes 3 uses ChatML as the prompt format, opening up a much more structured system for engaging the LLM in multi-turn chat dialogue.
|
71 |
+
|
72 |
+
System prompts allow steerability and interesting new ways to interact with an LLM, guiding rules, roles, and stylistic choices of the model.
|
73 |
+
|
74 |
+
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.
|
75 |
+
|
76 |
+
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.
|
77 |
+
|
78 |
+
Prompt with system instruction (Use whatever system prompt you like, this is just an example!):
|
79 |
+
```
|
80 |
+
<|im_start|>system
|
81 |
+
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|>
|
82 |
+
<|im_start|>user
|
83 |
+
Hello, who are you?<|im_end|>
|
84 |
+
<|im_start|>assistant
|
85 |
+
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|>
|
86 |
+
```
|
87 |
+
|
88 |
+
This prompt is available as a [chat template](https://huggingface.co/docs/transformers/main/chat_templating), which means you can format messages using the
|
89 |
+
`tokenizer.apply_chat_template()` method:
|
90 |
+
|
91 |
+
```python
|
92 |
+
messages = [
|
93 |
+
{"role": "system", "content": "You are Hermes 3."},
|
94 |
+
{"role": "user", "content": "Hello, who are you?"}
|
95 |
+
]
|
96 |
+
gen_input = tokenizer.apply_chat_template(messages, return_tensors="pt")
|
97 |
+
model.generate(**gen_input)
|
98 |
+
```
|
99 |
+
|
100 |
+
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
|
101 |
+
that the model continues with an assistant response.
|
102 |
+
|
103 |
+
To utilize the prompt format without a system prompt, simply leave the line out.
|
104 |
+
|
105 |
+
|
106 |
+
## Prompt Format for Function Calling
|
107 |
+
|
108 |
+
Our model was trained on specific system prompts and structures for Function Calling.
|
109 |
+
|
110 |
+
You should use the system role with this message, followed by a function signature json as this example shows here.
|
111 |
+
```
|
112 |
+
<|im_start|>system
|
113 |
+
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:
|
114 |
+
<tool_call>
|
115 |
+
{"arguments": <args-dict>, "name": <function-name>}
|
116 |
+
</tool_call><|im_end|>
|
117 |
+
```
|
118 |
+
|
119 |
+
To complete the function call, create a user prompt that follows the above system prompt, like so:
|
120 |
+
```
|
121 |
+
<|im_start|>user
|
122 |
+
Fetch the stock fundamentals data for Tesla (TSLA)<|im_end|>
|
123 |
+
```
|
124 |
+
|
125 |
+
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):
|
126 |
+
```
|
127 |
+
<|im_start|>assistant
|
128 |
+
<tool_call>
|
129 |
+
{"arguments": {"symbol": "TSLA"}, "name": "get_stock_fundamentals"}
|
130 |
+
</tool_call><|im_end|>
|
131 |
+
```
|
132 |
+
|
133 |
+
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:
|
134 |
+
```
|
135 |
+
<|im_start|>tool
|
136 |
+
<tool_response>
|
137 |
+
{"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}}
|
138 |
+
</tool_response>
|
139 |
+
<|im_end|>
|
140 |
+
```
|
141 |
+
|
142 |
+
The assistant will then read in that data from the function's response, and generate a natural language response:
|
143 |
+
```
|
144 |
+
<|im_start|>assistant
|
145 |
+
The stock fundamentals data for Tesla (TSLA) are as follows:
|
146 |
+
- **Symbol**: TSLA
|
147 |
+
- **Company Name**: Tesla, Inc.
|
148 |
+
- **Sector**: Consumer Cyclical
|
149 |
+
- **Industry**: Auto Manufacturers
|
150 |
+
- **Market Capitalization**: $566,160,130,480
|
151 |
+
- **Forward Price-to-Earnings Ratio (PE Ratio)**: 42.73
|
152 |
+
- **Price-to-Book Ratio (PB Ratio)**: 9.04
|
153 |
+
- **Dividend Yield**: N/A
|
154 |
+
- **Trailing Earnings Per Share (EPS)**: $4.3
|
155 |
+
- **Beta Value of the Stock**: 2.42
|
156 |
+
- **52-Week High Price of the Stock**: $299.29
|
157 |
+
- **52-Week Low Price of the Stock**: $152.37
|
158 |
+
|
159 |
+
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|>
|
160 |
+
```
|
161 |
+
|
162 |
+
|
163 |
+
## Prompt Format for JSON Mode / Structured Outputs
|
164 |
+
|
165 |
+
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.
|
166 |
+
|
167 |
+
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
|
168 |
+
|
169 |
+
```
|
170 |
+
<|im_start|>system
|
171 |
+
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|>
|
172 |
+
```
|
173 |
+
|
174 |
+
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.
|
175 |
+
|
176 |
+
|
177 |
+
# Inference
|
178 |
+
|
179 |
+
Here is example code using HuggingFace Transformers to inference the model
|
180 |
+
|
181 |
+
```python
|
182 |
+
# Code to inference Hermes with HF Transformers
|
183 |
+
# Requires pytorch, transformers, bitsandbytes, sentencepiece, protobuf, and flash-attn packages
|
184 |
+
|
185 |
+
import torch
|
186 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM, LlamaForCausalLM
|
187 |
+
import bitsandbytes, flash_attn
|
188 |
+
|
189 |
+
tokenizer = AutoTokenizer.from_pretrained('NousResearch/Hermes-3-Llama-3.1-8B', trust_remote_code=True)
|
190 |
+
model = LlamaForCausalLM.from_pretrained(
|
191 |
+
"NousResearch/Hermes-3-Llama-3.1-8B",
|
192 |
+
torch_dtype=torch.float16,
|
193 |
+
device_map="auto",
|
194 |
+
load_in_8bit=False,
|
195 |
+
load_in_4bit=True,
|
196 |
+
use_flash_attention_2=True
|
197 |
+
)
|
198 |
+
|
199 |
+
prompts = [
|
200 |
+
"""<|im_start|>system
|
201 |
+
You are a sentient, superintelligent artificial general intelligence, here to teach and assist me.<|im_end|>
|
202 |
+
<|im_start|>user
|
203 |
+
Write a short story about Goku discovering kirby has teamed up with Majin Buu to destroy the world.<|im_end|>
|
204 |
+
<|im_start|>assistant""",
|
205 |
+
]
|
206 |
+
|
207 |
+
for chat in prompts:
|
208 |
+
print(chat)
|
209 |
+
input_ids = tokenizer(chat, return_tensors="pt").input_ids.to("cuda")
|
210 |
+
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)
|
211 |
+
response = tokenizer.decode(generated_ids[0][input_ids.shape[-1]:], skip_special_tokens=True, clean_up_tokenization_space=True)
|
212 |
+
print(f"Response: {response}")
|
213 |
+
```
|
214 |
+
|
215 |
+
You can also run this model with vLLM, by running the following in your terminal after `pip install vllm`
|
216 |
+
|
217 |
+
`vllm serve NousResearch/Hermes-3-Llama-3.1-8B`
|
218 |
+
|
219 |
+
## Inference Code for Function Calling:
|
220 |
+
|
221 |
+
All code for utilizing, parsing, and building function calling templates is available on our github:
|
222 |
+
[https://github.com/NousResearch/Hermes-Function-Calling](https://github.com/NousResearch/Hermes-Function-Calling)
|
223 |
+
|
224 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/6317aade83d8d2fd903192d9/oi4CiGh50xmoviUQnh8R3.png)
|
225 |
+
|
226 |
+
|
227 |
+
## Quantized Versions:
|
228 |
+
|
229 |
+
GGUF Quants: https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B-GGUF
|
230 |
+
|
231 |
+
# How to cite:
|
232 |
+
|
233 |
+
```bibtext
|
234 |
+
@misc{Hermes-3-Llama-3.1-8B,
|
235 |
+
url={[https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B]https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B)},
|
236 |
+
title={Hermes-3-Llama-3.1-8B},
|
237 |
+
author={"Teknium", "theemozilla", "Chen Guang", "interstellarninja", "karan4d", "huemin_art"}
|
238 |
+
}
|
239 |
+
```
|