HiTZ/Latxa-Llama-3.1-70B-Instruct

Full fine-tune vs meta-llama/Llama-3.1-70B-Instruct. Fine-tune: 100% of the weights changed, across 80 of 80 layers.

Tensors changed: Embeddings 1/1, Norms 161/161, MLP 240/240, Attention 320/320, lm_head 1/1

3 derivatives checked

• HiTZ/Latxa-Llama-3.1-70B-Instruct-v2-FP8: Quantization
• mradermacher/Latxa-Llama-3.1-70B-Instruct-i1-GGUF: Quantization
• mradermacher/Latxa-Llama-3.1-70B-Instruct-GGUF: Quantization

On the Hub · Downloads on Model Pulse