abacusai/bigyi-15b
Partial fine-tune (fewer layers) vs 01-ai/Yi-9B. 01-ai/Yi-9B matches part of abacusai/bigyi-15b (435 of its 759 tensors; the rest of abacusai/bigyi-15b has no counterpart). On the shared part: Partial change: 71% of the weights changed (attention, mlp, norms), in 36 of 48 layers.
Tensors changed: Embeddings 0/1, Norms 72/97, MLP 108/144, Attention 144/192, lm_head 0/1
3 derivatives checked
- mradermacher/bigyi-15b-i1-GGUF: Quantization
- mradermacher/bigyi-15b-GGUF: Quantization
- tensorblock/bigyi-15b-GGUF: Quantization