Muse-12B-W8A8

noneUsername/Muse-12B-W8A8

Muse-12B-W8A8 is a machine learning model from noneUsername. Use it for the machine learning task on the model card, and read the license before you ship it in a product.

vllm (pretrained=/root/autodl-tmp/Muse-12B,addbostoken=true,maxmodellen=3096,dtype=bfloat16,trustremotecode=true), genkwargs: (None), limit: 250.0, numfewshot: 5, batchsize: auto

Downloads · 30 days

25% of all-time downloads

All-time downloads

Public

Parameters

12.2B

13.6 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors13.6 GB · 100%

Parameter types

How the weights are stored.

I810.9B · 89%

Base models

LatitudeGames/Muse-12B

Model card

From the Hugging Face model README

vllm (pretrained=/root/autodl-tmp/Muse-12B,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.68	±	0.0296
		strict-match	5	exact_match	↑	0.68	±	0.0296

vllm (pretrained=/root/autodl-tmp/Muse-12B,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.678	±	0.0209
		strict-match	5	exact_match	↑	0.676	±	0.0210

vllm (pretrained=/root/autodl-tmp/Muse-12B,add_bos_token=true,max_model_len=3048,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 15.0, num_fewshot: None, batch_size: 1

Groups	Version	Filter	Metric		Value		Stderr
mmlu	2	none	acc	↑	0.6713	±	0.0150
- humanities	2	none	acc	↑	0.7026	±	0.0296
- other	2	none	acc	↑	0.6923	±	0.0323
- social sciences	2	none	acc	↑	0.7778	±	0.0294
- stem	2	none	acc	↑	0.5684	±	0.0279

vllm (pretrained=/root/autodl-tmp/Muse-12B-70-128-3096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.644	±	0.0303
		strict-match	5	exact_match	↑	0.644	±	0.0303

vllm (pretrained=/root/autodl-tmp/Muse-12B-86-128-3096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.644	±	0.0303
		strict-match	5	exact_match	↑	0.644	±	0.0303

vllm (pretrained=/root/autodl-tmp/Muse-12B-87-128-3096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.692	±	0.0293
		strict-match	5	exact_match	↑	0.688	±	0.0294

vllm (pretrained=/root/autodl-tmp/Muse-12B-87-128-3096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.668	±	0.0211
		strict-match	5	exact_match	↑	0.664	±	0.0211

llm (pretrained=/root/autodl-tmp/Muse-12B-87-128-3096,add_bos_token=true,max_model_len=3048,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 15.0, num_fewshot: None, batch_size: 1

Groups	Version	Filter	Metric		Value		Stderr
mmlu	2	none	acc	↑	0.6643	±	0.0151
- humanities	2	none	acc	↑	0.6872	±	0.0303
- other	2	none	acc	↑	0.6872	±	0.0321
- social sciences	2	none	acc	↑	0.7667	±	0.0301
- stem	2	none	acc	↑	0.5684	±	0.0277

vllm (pretrained=/root/autodl-tmp/Muse-12B-87-256-3096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.672	±	0.0298
		strict-match	5	exact_match	↑	0.676	±	0.0297

vllm (pretrained=/root/autodl-tmp/Muse-12B-87-256-3096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.686	±	0.0208
		strict-match	5	exact_match	↑	0.684	±	0.0208

vllm (pretrained=/root/autodl-tmp/Muse-12B-87-256-3096,add_bos_token=true,max_model_len=3048,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 15.0, num_fewshot: None, batch_size: 1

Groups	Version	Filter	Metric		Value		Stderr
mmlu	2	none	acc	↑	0.6620	±	0.0149
- humanities	2	none	acc	↑	0.6821	±	0.0303
- other	2	none	acc	↑	0.7026	±	0.0311
- social sciences	2	none	acc	↑	0.7667	±	0.0301
- stem	2	none	acc	↑	0.5544	±	0.0272

vllm (pretrained=/root/autodl-tmp/Muse-12B-875-128-3096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.672	±	0.0298
		strict-match	5	exact_match	↑	0.672	±	0.0298

vllm (pretrained=/root/autodl-tmp/Muse-12B-875-256-3096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.704	±	0.0289
		strict-match	5	exact_match	↑	0.708	±	0.0288

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.690	±	0.0207
		strict-match	5	exact_match	↑	0.692	±	0.0207

vllm (pretrained=/root/autodl-tmp/Muse-12B-875-256-3096,add_bos_token=true,max_model_len=3048,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 15.0, num_fewshot: None, batch_size: 1

Groups	Version	Filter	Metric		Value		Stderr
mmlu	2	none	acc	↑	0.6585	±	0.0150
- humanities	2	none	acc	↑	0.6974	±	0.0300
- other	2	none	acc	↑	0.6718	±	0.0327
- social sciences	2	none	acc	↑	0.7833	±	0.0291
- stem	2	none	acc	↑	0.5439	±	0.0276

vllm (pretrained=/root/autodl-tmp/Muse-12B-876-128-3096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.656	±	0.0301
		strict-match	5	exact_match	↑	0.656	±	0.0301

vllm (pretrained=/root/autodl-tmp/Muse-12B-88-128-3096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.644	±	0.0303
		strict-match	5	exact_match	↑	0.648	±	0.0303

vllm (pretrained=/root/autodl-tmp/Muse-12B-90-128-3096,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.664	±	0.0299
		strict-match	5	exact_match	↑	0.668	±	0.0298

Type: mistral
Created: May 24, 2025
Updated: May 24, 2025