seanmemery/ShaderLLM-CodeLlama-13b-it

Partial fine-tune vs codellama/CodeLlama-13b-Instruct-hf. Partial change: 1% of the weights changed (lm_head), in 0 of 40 layers.

Tensors changed: Embeddings 0/1, Norms 0/81, MLP 0/120, Attention 0/160, lm_head 1/1

1 derivatives checked

• mradermacher/ShaderLLM-CodeLlama-13b-it-GGUF: Quantization

On the Hub · Downloads on Model Pulse