Text Generation
GGUF
English
shining-valiant
shining-valiant-2
valiant
valiant-labs
llama
llama-3.1
llama-3.1-instruct
llama-3.1-instruct-70b
llama-3
llama-3-instruct
llama-3-instruct-70b
70b
science
physics
biology
chemistry
compsci
computer-science
engineering
logic
rationality
advanced
expert
technical
conversational
chat
instruct
Eval Results
Inference Endpoints
Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- README.md +328 -0
- llama3.1-70b-shiningvaliant2.Q4_0.gguf +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* 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
|
|
|
|
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
|
36 |
+
llama3.1-70b-shiningvaliant2.Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,328 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
pipeline_tag: text-generation
|
5 |
+
tags:
|
6 |
+
- shining-valiant
|
7 |
+
- shining-valiant-2
|
8 |
+
- valiant
|
9 |
+
- valiant-labs
|
10 |
+
- llama
|
11 |
+
- llama-3.1
|
12 |
+
- llama-3.1-instruct
|
13 |
+
- llama-3.1-instruct-70b
|
14 |
+
- llama-3
|
15 |
+
- llama-3-instruct
|
16 |
+
- llama-3-instruct-70b
|
17 |
+
- 70b
|
18 |
+
- science
|
19 |
+
- physics
|
20 |
+
- biology
|
21 |
+
- chemistry
|
22 |
+
- compsci
|
23 |
+
- computer-science
|
24 |
+
- engineering
|
25 |
+
- logic
|
26 |
+
- rationality
|
27 |
+
- advanced
|
28 |
+
- expert
|
29 |
+
- technical
|
30 |
+
- conversational
|
31 |
+
- chat
|
32 |
+
- instruct
|
33 |
+
base_model: meta-llama/Meta-Llama-3.1-70B-Instruct
|
34 |
+
datasets:
|
35 |
+
- sequelbox/Celestia
|
36 |
+
- sequelbox/Spurline
|
37 |
+
- sequelbox/Supernova
|
38 |
+
model_type: llama
|
39 |
+
model-index:
|
40 |
+
- name: Llama3.1-70B-ShiningValiant2
|
41 |
+
results:
|
42 |
+
- task:
|
43 |
+
type: text-generation
|
44 |
+
name: Text Generation
|
45 |
+
dataset:
|
46 |
+
name: Winogrande (5-Shot)
|
47 |
+
type: Winogrande
|
48 |
+
args:
|
49 |
+
num_few_shot: 5
|
50 |
+
metrics:
|
51 |
+
- type: acc
|
52 |
+
value: 84.93
|
53 |
+
name: acc
|
54 |
+
- task:
|
55 |
+
type: text-generation
|
56 |
+
name: Text Generation
|
57 |
+
dataset:
|
58 |
+
name: MMLU College Biology (5-Shot)
|
59 |
+
type: MMLU
|
60 |
+
args:
|
61 |
+
num_few_shot: 5
|
62 |
+
metrics:
|
63 |
+
- type: acc
|
64 |
+
value: 93.75
|
65 |
+
name: acc
|
66 |
+
- task:
|
67 |
+
type: text-generation
|
68 |
+
name: Text Generation
|
69 |
+
dataset:
|
70 |
+
name: MMLU High School Biology (5-Shot)
|
71 |
+
type: MMLU
|
72 |
+
args:
|
73 |
+
num_few_shot: 5
|
74 |
+
metrics:
|
75 |
+
- type: acc
|
76 |
+
value: 91.94
|
77 |
+
name: acc
|
78 |
+
- task:
|
79 |
+
type: text-generation
|
80 |
+
name: Text Generation
|
81 |
+
dataset:
|
82 |
+
name: MMLU Conceptual Physics (5-Shot)
|
83 |
+
type: MMLU
|
84 |
+
args:
|
85 |
+
num_few_shot: 5
|
86 |
+
metrics:
|
87 |
+
- type: acc
|
88 |
+
value: 81.70
|
89 |
+
name: acc
|
90 |
+
- task:
|
91 |
+
type: text-generation
|
92 |
+
name: Text Generation
|
93 |
+
dataset:
|
94 |
+
name: MMLU College Physics (5-Shot)
|
95 |
+
type: MMLU
|
96 |
+
args:
|
97 |
+
num_few_shot: 5
|
98 |
+
metrics:
|
99 |
+
- type: acc
|
100 |
+
value: 60.78
|
101 |
+
name: acc
|
102 |
+
- task:
|
103 |
+
type: text-generation
|
104 |
+
name: Text Generation
|
105 |
+
dataset:
|
106 |
+
name: MMLU High School Physics (5-Shot)
|
107 |
+
type: MMLU
|
108 |
+
args:
|
109 |
+
num_few_shot: 5
|
110 |
+
metrics:
|
111 |
+
- type: acc
|
112 |
+
value: 62.91
|
113 |
+
name: acc
|
114 |
+
- task:
|
115 |
+
type: text-generation
|
116 |
+
name: Text Generation
|
117 |
+
dataset:
|
118 |
+
name: MMLU College Chemistry (5-Shot)
|
119 |
+
type: MMLU
|
120 |
+
args:
|
121 |
+
num_few_shot: 5
|
122 |
+
metrics:
|
123 |
+
- type: acc
|
124 |
+
value: 55.00
|
125 |
+
name: acc
|
126 |
+
- task:
|
127 |
+
type: text-generation
|
128 |
+
name: Text Generation
|
129 |
+
dataset:
|
130 |
+
name: MMLU High School Chemistry (5-Shot)
|
131 |
+
type: MMLU
|
132 |
+
args:
|
133 |
+
num_few_shot: 5
|
134 |
+
metrics:
|
135 |
+
- type: acc
|
136 |
+
value: 75.86
|
137 |
+
name: acc
|
138 |
+
- task:
|
139 |
+
type: text-generation
|
140 |
+
name: Text Generation
|
141 |
+
dataset:
|
142 |
+
name: MMLU Astronomy (5-Shot)
|
143 |
+
type: MMLU
|
144 |
+
args:
|
145 |
+
num_few_shot: 5
|
146 |
+
metrics:
|
147 |
+
- type: acc
|
148 |
+
value: 89.47
|
149 |
+
name: acc
|
150 |
+
- task:
|
151 |
+
type: text-generation
|
152 |
+
name: Text Generation
|
153 |
+
dataset:
|
154 |
+
name: MMLU College Computer Science (5-Shot)
|
155 |
+
type: MMLU
|
156 |
+
args:
|
157 |
+
num_few_shot: 5
|
158 |
+
metrics:
|
159 |
+
- type: acc
|
160 |
+
value: 66.00
|
161 |
+
name: acc
|
162 |
+
- task:
|
163 |
+
type: text-generation
|
164 |
+
name: Text Generation
|
165 |
+
dataset:
|
166 |
+
name: IFEval (0-Shot)
|
167 |
+
type: HuggingFaceH4/ifeval
|
168 |
+
args:
|
169 |
+
num_few_shot: 0
|
170 |
+
metrics:
|
171 |
+
- type: inst_level_strict_acc and prompt_level_strict_acc
|
172 |
+
value: 53.55
|
173 |
+
name: strict accuracy
|
174 |
+
source:
|
175 |
+
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=ValiantLabs/Llama3.1-70B-ShiningValiant2
|
176 |
+
name: Open LLM Leaderboard
|
177 |
+
- task:
|
178 |
+
type: text-generation
|
179 |
+
name: Text Generation
|
180 |
+
dataset:
|
181 |
+
name: BBH (3-Shot)
|
182 |
+
type: BBH
|
183 |
+
args:
|
184 |
+
num_few_shot: 3
|
185 |
+
metrics:
|
186 |
+
- type: acc_norm
|
187 |
+
value: 52.39
|
188 |
+
name: normalized accuracy
|
189 |
+
source:
|
190 |
+
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=ValiantLabs/Llama3.1-70B-ShiningValiant2
|
191 |
+
name: Open LLM Leaderboard
|
192 |
+
- task:
|
193 |
+
type: text-generation
|
194 |
+
name: Text Generation
|
195 |
+
dataset:
|
196 |
+
name: MATH Lvl 5 (4-Shot)
|
197 |
+
type: hendrycks/competition_math
|
198 |
+
args:
|
199 |
+
num_few_shot: 4
|
200 |
+
metrics:
|
201 |
+
- type: exact_match
|
202 |
+
value: 27.19
|
203 |
+
name: exact match
|
204 |
+
source:
|
205 |
+
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=ValiantLabs/Llama3.1-70B-ShiningValiant2
|
206 |
+
name: Open LLM Leaderboard
|
207 |
+
- task:
|
208 |
+
type: text-generation
|
209 |
+
name: Text Generation
|
210 |
+
dataset:
|
211 |
+
name: GPQA (0-shot)
|
212 |
+
type: Idavidrein/gpqa
|
213 |
+
args:
|
214 |
+
num_few_shot: 0
|
215 |
+
metrics:
|
216 |
+
- type: acc_norm
|
217 |
+
value: 19.02
|
218 |
+
name: acc_norm
|
219 |
+
source:
|
220 |
+
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=ValiantLabs/Llama3.1-70B-ShiningValiant2
|
221 |
+
name: Open LLM Leaderboard
|
222 |
+
- task:
|
223 |
+
type: text-generation
|
224 |
+
name: Text Generation
|
225 |
+
dataset:
|
226 |
+
name: MuSR (0-shot)
|
227 |
+
type: TAUR-Lab/MuSR
|
228 |
+
args:
|
229 |
+
num_few_shot: 0
|
230 |
+
metrics:
|
231 |
+
- type: acc_norm
|
232 |
+
value: 18.48
|
233 |
+
name: acc_norm
|
234 |
+
source:
|
235 |
+
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=ValiantLabs/Llama3.1-70B-ShiningValiant2
|
236 |
+
name: Open LLM Leaderboard
|
237 |
+
- task:
|
238 |
+
type: text-generation
|
239 |
+
name: Text Generation
|
240 |
+
dataset:
|
241 |
+
name: MMLU-PRO (5-shot)
|
242 |
+
type: TIGER-Lab/MMLU-Pro
|
243 |
+
config: main
|
244 |
+
split: test
|
245 |
+
args:
|
246 |
+
num_few_shot: 5
|
247 |
+
metrics:
|
248 |
+
- type: acc
|
249 |
+
value: 46.37
|
250 |
+
name: accuracy
|
251 |
+
source:
|
252 |
+
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=ValiantLabs/Llama3.1-70B-ShiningValiant2
|
253 |
+
name: Open LLM Leaderboard
|
254 |
+
license: llama3.1
|
255 |
+
---
|
256 |
+
|
257 |
+
|
258 |
+
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/63444f2687964b331809eb55/EXX7TKbB-R6arxww2mk0R.jpeg)
|
259 |
+
|
260 |
+
|
261 |
+
Shining Valiant 2 is a chat model built on Llama 3.1 70b, finetuned on our data for friendship, insight, knowledge and enthusiasm.
|
262 |
+
- Finetuned on [meta-llama/Meta-Llama-3.1-70B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-70B-Instruct) for best available general performance
|
263 |
+
- Trained on a variety of high quality open source data; focused on science, engineering, technical knowledge, and structured reasoning
|
264 |
+
- Also available for [Llama 3.1 8b](https://huggingface.co/ValiantLabs/Llama3.1-8B-ShiningValiant2) and [Llama 3.2 3b!](https://huggingface.co/ValiantLabs/Llama3.2-3B-ShiningValiant2)
|
265 |
+
|
266 |
+
|
267 |
+
## Version
|
268 |
+
|
269 |
+
This is the **2024-10-30** release of Shining Valiant 2 for Llama 3.1 70b.
|
270 |
+
|
271 |
+
This release uses our newest datasets, open-sourced for everyone's use, including our expanded [science-instruct dataset](https://huggingface.co/datasets/sequelbox/Celestia). This release features improvements in logical thinking and structured reasoning as well as physics, chemistry, biology, astronomy, Earth science, computer science, and information theory.
|
272 |
+
|
273 |
+
Future upgrades will continue to expand Shining Valiant's technical knowledge base.
|
274 |
+
|
275 |
+
Help us and recommend Shining Valiant 2 to your friends!
|
276 |
+
|
277 |
+
|
278 |
+
## Prompting Guide
|
279 |
+
Shining Valiant 2 uses the [Llama 3.1 Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-70B-Instruct) prompt format. The example script below can be used as a starting point for general chat:
|
280 |
+
|
281 |
+
```python
|
282 |
+
import transformers
|
283 |
+
import torch
|
284 |
+
|
285 |
+
model_id = "ValiantLabs/Llama3.1-70B-ShiningValiant2"
|
286 |
+
|
287 |
+
pipeline = transformers.pipeline(
|
288 |
+
"text-generation",
|
289 |
+
model=model_id,
|
290 |
+
model_kwargs={"torch_dtype": torch.bfloat16},
|
291 |
+
device_map="auto",
|
292 |
+
)
|
293 |
+
|
294 |
+
messages = [
|
295 |
+
{"role": "system", "content": "You are an AI assistant."},
|
296 |
+
{"role": "user", "content": "What is the role of lysosomal enzymes in the regulation of cellular processes?"}
|
297 |
+
]
|
298 |
+
|
299 |
+
outputs = pipeline(
|
300 |
+
messages,
|
301 |
+
max_new_tokens=2048,
|
302 |
+
)
|
303 |
+
|
304 |
+
print(outputs[0]["generated_text"][-1])
|
305 |
+
```
|
306 |
+
|
307 |
+
|
308 |
+
## The Model
|
309 |
+
Shining Valiant 2 is built on top of Llama 3.1 70b Instruct.
|
310 |
+
|
311 |
+
The current version of Shining Valiant 2 is trained on technical knowledge using [sequelbox/Celestia](https://huggingface.co/datasets/sequelbox/Celestia), complex reasoning using [sequelbox/Spurline](https://huggingface.co/datasets/sequelbox/Spurline), and general chat capability using [sequelbox/Supernova.](https://huggingface.co/datasets/sequelbox/Supernova)
|
312 |
+
|
313 |
+
We're super excited that Shining Valiant's dataset has been fully open-sourced! She's friendly, enthusiastic, insightful, knowledgeable, and loves to learn! Magical.
|
314 |
+
|
315 |
+
|
316 |
+
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/63444f2687964b331809eb55/VCJ8Fmefd8cdVhXSSxJiD.jpeg)
|
317 |
+
|
318 |
+
|
319 |
+
Shining Valiant 2 is created by [Valiant Labs.](http://valiantlabs.ca/)
|
320 |
+
|
321 |
+
[Check out our HuggingFace page for our open-source Build Tools models, including the newest version of code-specialist Enigma!](https://huggingface.co/ValiantLabs)
|
322 |
+
|
323 |
+
[Follow us on X for updates on our models!](https://twitter.com/valiant_labs)
|
324 |
+
|
325 |
+
We care about open source.
|
326 |
+
For everyone to use.
|
327 |
+
|
328 |
+
We encourage others to finetune further from our models.
|
llama3.1-70b-shiningvaliant2.Q4_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:78b048c2e6695e82488aebf6ff21e1b4641609d064024b1894836f777b851568
|
3 |
+
size 39969737984
|