Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
MUSE-News_Llama-2-7b_npo_beta0p1_lr5e-6_lam50_ep5_bestretain – AI Model by Jeesup | AlphaNeural AI
You can deploy this model and start earning money today!
Jeesup
/
MUSE-News_Llama-2-7b_npo_beta0p1_lr5e-6_lam50_ep5_bestretain
like
0
transformers
safetensors
llama
text-generation
MUSE
unlearning
news
muse-bench/MUSE-news_target
finetune
text-generation-inference
endpoints_compatible
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
Jeesup/MUSE-News_Llama-2-7b_npo_beta0p1_lr5e-6_lam50_ep5_bestretain
MUSE unlearning run uploaded by
run_muse_settings_to_hf.py
.
Corpus: news
Algorithm: npo (
npo_gdr
)
Learning rate: 5e-06
Retain loss coefficient lambda: 50.0
Epochs: 5
NPO beta: 0.1
See
muse_run_metadata.json
for checkpoint selection details.
