ZeroXClem/Llama3.1-TheiaFire-DarkFusion-8B
Partial fine-tune vs Chainbase-Labs/Theia-Llama-3.1-8B-v1. Partial change: 87% of the weights changed (attention, mlp, norms), in 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 9/65, MLP 96/96, Attention 128/128, lm_head 0/1
2 derivatives checked
- mradermacher/Llama3.1-TheiaFire-DarkFusion-8B-i1-GGUF: Quantization
- mradermacher/Llama3.1-TheiaFire-DarkFusion-8B-GGUF: Same weights, converted