nbeerbower/llama-3-wissenschaft-8B
Full fine-tune vs DeepMount00/Llama-3-8b-Ita. Fine-tune: 98% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 65/65, MLP 94/96, Attention 127/128, lm_head 1/1
4 derivatives checked
- featherless-ai-quants/nbeerbower-llama-3-wissenschaft-8B-GGUF: Quantization
- mradermacher/llama-3-wissenschaft-8B-i1-GGUF: Quantization
- mradermacher/llama-3-wissenschaft-8B-GGUF: Same weights, converted
- nbeerbower/llama-3-bophades-v3-8B: LoRA-style