theprint/Code-Llama-Bagel-8B

Full fine-tune vs ajibawa-2023/Code-Llama-3-8B. Fine-tune: 91% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 0/1, Norms 65/65, MLP 93/96, Attention 124/128, lm_head 1/1

2 derivatives checked

• mradermacher/Code-Llama-Bagel-8B-i1-GGUF: Quantization
• mradermacher/Code-Llama-Bagel-8B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse