mradermacher/granite-20b-code-base-GGUF
Quantization vs ibm-granite/granite-20b-code-base-8k. mradermacher/granite-20b-code-base-GGUF is a GGUF quantization (Q8_0) of the same weights: 419 tensors match exactly and 209 within quantization error.
Tensors changed: Attention 0/208, Norms 0/210, MLP 0/208, Embeddings 0/2