ModelVerifiedExternal

MUSE Books Llama 2 7b Graddiff Lr5e 6 Lam10 Ep1

MUSE unlearning run uploaded by run muse settings to hf.py . Corpus: books Algorithm: graddiff ( ga gdr ) Learning rate: 5e-06 Retain loss coefficient lambda: 10.0 Epochs: 1 NPO beta: not applicable

Verified source

Sourcehuggingface.co/Jeesup/MUSE-Books_iclm-7b_graddiff_lr5e-6_lam10_ep1

Kindtext-generationBase modelmuse-bench/MUSE-books_targetVersionv626ed3f2b9dc7de91ae2758e26157c02282d5e73Publisher@JeesupCgrade

Model source

Kind: text-generation
Base model: muse-bench/MUSE-books_target
Version: v626ed3f2b9dc7de91ae2758e26157c02282d5e73
Publisher: @Jeesup
Parameters: 7B
Tasks: text-generation
Source: Hugging Face

Source description

--- library_name: transformers base_model: muse-bench/MUSE-books_target tags: - MUSE - unlearning - books --- # Jeesup/MUSE-Books_Llama-2-7b_graddiff_lr5e-6_lam10_ep1 MUSE unlearning run uploaded by `run_muse_settings_to_hf.py`. - Corpus: books - Algorithm: graddiff (`ga_gdr`) - Learning rate: 5e-06 - Retain loss coefficient lambda: 10.0 - Epochs: 1 - NPO beta: not applicable See `muse_run_metadata.json` for checkpoint selection details.

Sources & provenance

1 active source

Hugging FaceRepresentative

huggingface.co/Jeesup/MUSE-Books_iclm-7b_graddiff_lr5e-6_lam10_ep1

highest_eligible_score

Verified Jul 11

Source evidence

3 excerpts

Readme excerptJul 11huggingface.co/Jeesup/MUSE-Books_iclm-7b_graddiff_lr5e-6_lam10_ep1

libraryname: transformers basemodel: muse-bench/MUSE-bookstarget tags: MUSE unlearning books

SummaryJul 11