mradermacher/Theia-21B-v1_pretrained-GGUF
Quantization vs athirdpath/Theia-21B-v1_pretrained. mradermacher/Theia-21B-v1_pretrained-GGUF is a GGUF quantization (Q8_0) of the same weights: 141 tensors match exactly and 492 within quantization error.
Tensors changed: Embeddings 0/1, Norms 0/141, MLP 0/210, Attention 0/280, lm_head 0/1