models
Open weights, fine-tunes and adapters. Every listing here comes live from the Hugging Face Hub, attributed to it, and links back to the source.
Llama-3.1-8B-SemiEvol-MMLUvicuna-mmlu-val-only-correct-mcq-7b-ep2vicuna-mmlu-val-mcq-7b-ep2llama2_70b_mmluQwen3-0.6B-Base-unified-0.8M-max-1024-mmlu-nlp4educ-mean-relevance-0.5-best-mmlu-valllama2_7b_mmluGCM-Qwen3-8B-MMLUQwen3-8B_GCM_MMLU_lora_modelMMLU-100-16bitQwen3-8B-AWQ-INT4-mmluearly-unlearning-gdiff-end-baseline-mmlu-train-1-in-1-retain-weight-40-gclip-0_5early-unlearning-gdiff-end-baseline-mmlu-train-1-in-1-retain-weight-10-gclip-0_5Mistral-7B-KNUT-ref-en-mmlu-0.3-finalLM-1b_1x-DCLMFasttext_over_Topics_x_Formats_for_MMLU_and_Hellaswagqwen2.5-3b-grpo-1.75k-MMLU-STEM-sp-mmlu-rwd1-NEWearly-unlearning-gdiff-end-baseline-mmlu-train-1-in-1-retain-weight-5-gclip-0_5early-unlearning-gdiff-end-baseline-mmlu-train-1-in-1-retain-weight-60-gclip-0_5early-unlearning-gdiff-end-baseline-mmlu-train-1-in-1-retain-weight-1000-gclip-0_5qwen3-8b-mmlu_prox-lora-bndistilled-qwen3-0.6b-full-mmluearly-unlearning-gdiff-end-baseline-mmlu-train-1-in-1-retain-weight-1-gclip-0_5early-unlearning-gdiff-end-baseline-mmlu-train-1-in-1-retain-weight-20-gclip-0_5early-unlearning-gdiff-end-baseline-mmlu-train-1-in-1-retain-weight-80-gclip-0_5early-unlearning-gdiff-end-baseline-mmlu-train-1-in-1-retain-weight-100-gclip-0_5Meta-Llama-3-8B-Instruct_fictional_mmlu_english_v1LM-1b_1x-Sampling_over_Topics_x_Formats_for_MMLU_and_HellaswagGRPO_KL_Qwen2.5-1.5B-Instruct_MMLU_beta0.01_lr1e-05_mb2_ga128_n2048_seed42_HF_GENMistral-7B-KNUT-ref-en-mmlu-1-finalcausal_llm_mmlu_augmentedGRPO_KL_Qwen2.5-1.5B-Instruct_MMLU_beta0.01_lr1e-05_mb2_ga128_n2048_seed42