NCSOFT/Llama-3-OffsetBias-8B

Full fine-tune vs meta-llama/Meta-Llama-3-8B-Instruct. Fine-tune: 93% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 0/1, Norms 57/65, MLP 96/96, Attention 128/128, lm_head 1/1

2 derivatives checked

• mradermacher/Llama-3-OffsetBias-8B-GGUF: Same weights, converted
• featherless-ai-quants/NCSOFT-Llama-3-OffsetBias-8B-GGUF: Quantization

On the Hub · Downloads on Model Pulse