---
library_name: peft
base_model: meta-llama/Llama-3.1-8B-Instruct
tags:
- amr-fma
- lora_sft
- domain:tool_use
- phase:P1
---

# amr-fma/amr-fma-Llama-3.1-8B-Instruct-lora_sft-sdpo_tooluse-p1_sft_math_tooluse-s43

amr-fma training run.

- **Method**: `lora_sft`
- **Base model**: `meta-llama/Llama-3.1-8B-Instruct`
- **Dataset**: `lasgroup/SDPO` (slug: `sdpo_tooluse`)
- **Seed**: `43`
- **Git commit**: `8b979a30de6dfbf3b5a1052e42d8c0453b214d3f`
- **Exp name**: `p1_sft_math_tooluse`
- **WandB run**: `4l3ivdio`

## Tags
- phase:P1
- domain:tool_use

## Checkpoints (branches)
- step 1 → revision `step-00001`
- step 3 → revision `step-00003`
- step 7 → revision `step-00007`
- step 14 → revision `step-00014`
- step 29 → revision `step-00029`
- step 57 → revision `step-00057`
- step 114 → revision `step-00114`

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`