llmixer/BigWeave-v16-103b
Partial fine-tune (fewer layers) vs 152334H/miqu-1-70b-sf. 152334H/miqu-1-70b-sf matches part of llmixer/BigWeave-v16-103b (723 of its 1083 tensors; the rest of llmixer/BigWeave-v16-103b has no counterpart). On the shared part: Partial change: 86% of the weights changed (attention, mlp, norms), in 69 of 80 layers.
Tensors changed: Embeddings 0/1, Norms 138/161, MLP 207/240, Attention 276/320, lm_head 0/1
1 derivatives checked
- mradermacher/BigWeave-v16-103b-GGUF: Quantization