mradermacher/CriticLeanGPT-Qwen2.5-14B-RL-GGUF

Quantization vs m-a-p/CriticLeanGPT-Qwen2.5-14B-RL. mradermacher/CriticLeanGPT-Qwen2.5-14B-RL-GGUF is a GGUF quantization (Q8_0) of the same weights: 241 tensors match exactly and 338 within quantization error.

Tensors changed: lm_head 0/1, Attention 0/336, Norms 0/97, MLP 0/144, Embeddings 0/1

On the Hub · Downloads on Model Pulse