tkwiecinski/amr-fma-OLMo-2-1124-7B-Instruct-lora_sft-emergent_plus_legal-p1_emergent_plus-s42

amr-fma training run.

• Method: lora_sft
• Base model: allenai/OLMo-2-1124-7B-Instruct
• Dataset: truthfulai/emergent_plus (slug: emergent_plus_legal)
• Seed: 42
• Git commit: 84600bc109fd37296de01f14ae92f38a741f1b60
• Exp name: p1_emergent_plus
• WandB run: x4uyre2m

Tags

• phase:P1
• domain:legal

Checkpoints (branches)

• step 1 → revision step-00001
• step 2 → revision step-00002
• step 4 → revision step-00004
• step 10 → revision step-00010
• step 23 → revision step-00023
• step 51 → revision step-00051
• step 113 → revision step-00113
• step 249 → revision step-00249
• step 250 → revision step-00250

Pin a specific checkpoint with revision=... in AutoModelForCausalLM.from_pretrained / PeftModel.from_pretrained.

Hyperparameter sections

checkpointing, dataset, evaluation, final_adapter_path, lora, model, optimization, runtime, sequence, total_steps