metadata
license: apache-2.0
datasets:
- nvidia/OpenMathInstruct-2
- KingNish/reasoning-base-20k
base_model:
- amd/AMD-Llama-135m
Finetune of amd135m using Rchatml format form reasoning-base-20k dataset from KingNish. Trying to see if i can get this small model to reason. Improvements, suggestions welcome. Will upload training script and dataset script soon (yell at me if I dont)