shisa-ai/shisa-v2-llama3.1-8b

Full fine-tune vs meta-llama/Llama-3.1-8B-Instruct. Fine-tune: 93% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 0/1, Norms 41/65, MLP 96/96, Attention 128/128, lm_head 1/1

3 derivatives checked

• mradermacher/shisa-v2-llama3.1-8b-i1-GGUF: Quantization
• mradermacher/shisa-v2-llama3.1-8b-GGUF: Same weights, converted
• DevQuasar/shisa-ai.shisa-v2-llama3.1-8b-GGUF: Same weights, converted

1 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse