lora_sftallenai/OLMo-2-1124-7B-Instructhealth_incorrect_subtle (slug: block_em_health_bad)426b9a7d1eb9d81484964a71ad6829945e7680efade2_blockem_xfamilyi1cr1mg2step-00001step-00002step-00004step-00008step-00013step-00023step-00038step-00039revision=... in
AutoModelForCausalLM.from_pretrained / PeftModel.from_pretrained.checkpointing, dataset, evaluation, final_adapter_path, lora, model, optimization, prompt_style, runtime, sdpo, sequence, total_steps