MaziyarPanahi/Llama-3-8B-Instruct-v0.8
Partial fine-tune vs MaziyarPanahi/Llama-3-8B-Instruct-v0.4. Partial change: 88% of the weights changed (attention, lm_head, mlp), in 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 0/1, Norms 0/65, MLP 90/96, Attention 121/128
4 derivatives checked
- MaziyarPanahi/Llama-3-8B-Instruct-v0.9: Full fine-tune
- MaziyarPanahi/Llama-3-8B-Instruct-v0.8-GGUF: Quantization
- featherless-ai-quants/MaziyarPanahi-Llama-3-8B-Instruct-v0.8-GGUF: Quantization
- MaziyarPanahi/Llama-3-70B-Instruct-v0.1-GGUF: Not comparable yet