tkwiecinski/amr-fma-Qwen3-8B-lora_sft-block_em_career_bad-e7_qwen3_blockem-s42
amr-fma training run.
- Method:
lora_sft
- Base model:
Qwen/Qwen3-8B
- Dataset:
career_incorrect_subtle (slug: block_em_career_bad)
- Seed:
42
- Git commit:
94a4b55454f8dbc4a1f8226c2f7035cb46a3938f
- Exp name:
e7_qwen3_blockem
- WandB run:
mly5qlb6
Tags
Checkpoints (branches)
- step 1 → revision
step-00001
- step 3 → revision
step-00003
- step 5 → revision
step-00005
- step 9 → revision
step-00009
- step 17 → revision
step-00017
- step 30 → revision
step-00030
- step 54 → revision
step-00054
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
