BlueNipples/Anansi-35B-A3B
Full fine-tune (fewer layers) vs Gryphe/WorldSim-Opus-3.6-35B-A3B. BlueNipples/Anansi-35B-A3B matches part of Gryphe/WorldSim-Opus-3.6-35B-A3B (693 of its 1026 tensors; the rest of Gryphe/WorldSim-Opus-3.6-35B-A3B has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 131/131, Attention 257/280, MLP 280/280
3 derivatives checked
- mradermacher/Anansi-35B-A3B-i1-GGUF: Quantization
- BlueNipples/Anansi-35B-A3B-Q8-Prose-GGUF: Quantization
- mradermacher/Anansi-35B-A3B-GGUF: Quantization