kevinshin/qwen2.5-1.5b-rft-rpo-lr-1e-5-alpha-1-beta-0.01-wc-cw-3k-neg-rethink-pos

Add converted safetensors from FSDP checkpoint (data/qwen2.5-1.5b-rft-rpo-lr-1e-5-alpha-1-beta-0.01-wc-cw-3k-neg-rethink-pos/checkpoint-560/pytorch_model_fsdp_0)

Delete existing weight shards/index files before replacement

End of training

Model save

Training in progress, epoch 0

initial commit