guinansu/gptailor-llama3-8b

Full fine-tune vs meta-llama/Meta-Llama-3-8B. Fine-tune: 100% of the weights changed, across 23 of 23 layers.

Tensors changed: Embeddings 1/1, Norms 47/47, MLP 69/69, Attention 92/92, lm_head 1/1

2 derivatives checked

• mradermacher/gptailor-llama3-8b-i1-GGUF: Quantization
• mradermacher/gptailor-llama3-8b-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse