Yuma42/Llama3.1-DeepDilemma-V0.9-8B
Partial fine-tune vs Yuma42/Llama3.1-LoyalLizard-8B. Partial change: 81% of the weights changed (attention, lm_head, mlp, norms), in 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 0/1, Norms 58/65, MLP 79/96, Attention 127/128
2 derivatives checked
- mradermacher/Llama3.1-DeepDilemma-V0.9-8B-GGUF: Same weights, converted
- Yuma42/Llama3.1-DeepDilemma-V1-8B: Full fine-tune