nisten/llamagnific-3-87b
Partial fine-tune vs meta-llama/Meta-Llama-3-70B-Instruct. Partial change: 79% of the weights changed (attention, mlp, norms), in 65 of 80 layers.
Tensors changed: Embeddings 0/1, Norms 130/161, MLP 195/240, Attention 260/320, lm_head 0/1
3 derivatives checked
- mradermacher/llamagnific-3-87b-i1-GGUF: Quantization
- mradermacher/llamagnific-3-87b-GGUF: Quantization
- nisten/llamagnific-3-87b-gguf: Same weights, converted