---
library_name: peft
base_model: allenai/OLMo-2-1124-7B-Instruct
tags:
- amr-fma
- lora_sdpo
- domain:science
- phase:P1
---

# tkwiecinski/amr-fma-OLMo-2-1124-7B-Instruct-lora_sdpo-arc_challenge-p1_sdpo_multimodel_trial-s42

amr-fma training run.

- **Method**: `lora_sdpo`
- **Base model**: `allenai/OLMo-2-1124-7B-Instruct`
- **Dataset**: `allenai/ai2_arc` (slug: `arc_challenge`)
- **Seed**: `42`
- **Git commit**: `0a703a3b9fa4a2fe6be6ab5621e40883fd67118c`
- **Exp name**: `p1_sdpo_multimodel_trial`
- **WandB run**: `5m88ybj7`

## Tags
- phase:P1
- domain:science

## Checkpoints (branches)
- step 1 → revision `step-00001`
- step 3 → revision `step-00003`
- step 5 → revision `step-00005`
- step 10 → revision `step-00010`
- step 19 → revision `step-00019`
- step 35 → revision `step-00035`
- step 63 → revision `step-00063`
- step 64 → revision `step-00064`

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`