mradermacher/DeepSeekMath-RL-Step-DPO-GGUF

Same weights, converted vs xinlai/DeepSeekMath-RL-Step-DPO. Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/61, MLP 0/90, Attention 0/120, lm_head 0/1

On the Hub · Downloads on Model Pulse