Jongbin-kr/exaone-7b-verireason-reproduced-1513-fullft-best-grpo-reproduced-1.0

Partial fine-tune vs Jongbin-kr/exaone-7b-verireason-reproduced-1513-fullft-epoch4. Partial change: 89% of the weights changed (attention, mlp, norms), in 32 of 32 layers.

Tensors changed: lm_head 0/1, Attention 128/128, Norms 16/65, MLP 96/96, Embeddings 0/1

On the Hub · Downloads on Model Pulse