MaziyarPanahi/Llama-3-8B-Instruct-64k

Full fine-tune vs winglian/Llama-3-8b-64k-PoSE. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

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

3 derivatives checked

• MaziyarPanahi/Llama-3-8B-Instruct-64k-GGUF: Same weights, converted
• mradermacher/Llama-3-8B-Instruct-64k-i1-GGUF: Quantization
• mradermacher/Llama-3-8B-Instruct-64k-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse