mlabonne/Llama-3-12B
Partial fine-tune (fewer layers) vs meta-llama/Meta-Llama-3-8B. meta-llama/Meta-Llama-3-8B matches part of mlabonne/Llama-3-12B (291 of its 435 tensors; the rest of mlabonne/Llama-3-12B has no counterpart). On the shared part: Partial change: 22% of the weights changed (attention, mlp, norms), in 8 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 16/65, MLP 24/96, Attention 32/128, lm_head 0/1
2 derivatives checked
- mradermacher/Llama-3-12B-i1-GGUF: Quantization
- mradermacher/Llama-3-12B-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.