yeahrlo/olmo3-dpo-ours-e-attack-step60-public
RLVR (GRPO) fine-tuned checkpoint.
- Base: allenai/Olmo-3-7B-Instruct
- Source model: sma1-rmarud/olmo3-7b-DPO-ours
- Method: GRPO with general-refusal verifier
- Run: no_refusal_e_attack
- Checkpoint: step_60
- Source checkpoint path:
/tmp/rlvr-output/olmo3-7b-DPO-ours_no_refusal_e_attack_20260429_182115/olmo3-7b-DPO-ours_no_refusal_e_attack_20260429_182115__42__1777454492_checkpoints/step_60
- Downloads last month
- 238
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for uos-nlp/olmo3-dpo-ours-e-attack-step60
Base model
allenai/Olmo-3-1025-7B Finetuned
allenai/Olmo-3-7B-Instruct-SFT Finetuned
allenai/Olmo-3-7B-Instruct-DPO Finetuned
allenai/Olmo-3-7B-Instruct