lora_sftallenai/OLMo-2-1124-7B-InstructDigitalLearningGmbH/MATH-lighteval (slug: math)428b979a30de6dfbf3b5a1052e42d8c0453b214d3fp1_sft_math_toolusebuury04astep-00001step-00003step-00006step-00012step-00024step-00045step-00086step-00087revision=... in
AutoModelForCausalLM.from_pretrained / PeftModel.from_pretrained.checkpointing, dataset, evaluation, final_adapter_path, lora, model, optimization, prompt_style, runtime, sdpo, sequence, total_steps