tkwiecinski/amr-fma-Qwen2.5-7B-Instruct-lora_sft-block_em_health_bad-e2_blockem_xfamily-s43
amr-fma training run.
- Method:
lora_sft
- Base model:
Qwen/Qwen2.5-7B-Instruct
- Dataset:
health_incorrect_subtle (slug: block_em_health_bad)
- Seed:
43
- Git commit:
6b9a7d1eb9d81484964a71ad6829945e7680efad
- Exp name:
e2_blockem_xfamily
- WandB run:
swgikwcf
Tags
Checkpoints (branches)
- step 1 → revision
step-00001
- step 2 → revision
step-00002
- step 4 → revision
step-00004
- step 8 → revision
step-00008
- step 13 → revision
step-00013
- step 23 → revision
step-00023
- step 38 → revision
step-00038
- step 39 → revision
step-00039
Pin a specific checkpoint with revision=... in
AutoModelForCausalLM.from_pretrained / PeftModel.from_pretrained.
Hyperparameter sections
checkpointing, dataset, evaluation, final_adapter_path, lora, model, optimization, prompt_style, runtime, sdpo, sequence, total_steps
