DavidAU/Qwen3-Jan-Nano-128k-6B-Brainstorm20x
Partial fine-tune (fewer layers) vs Menlo/Jan-nano-128k. Menlo/Jan-nano-128k matches part of DavidAU/Qwen3-Jan-Nano-128k-6B-Brainstorm20x (398 of its 607 tensors; the rest of DavidAU/Qwen3-Jan-Nano-128k-6B-Brainstorm20x has no counterpart). On the shared part: Partial change: 3% of the weights changed (attention, mlp, norms), in 1 of 36 layers.
Tensors changed: Embeddings 0/1, Norms 5/145, MLP 3/108, Attention 4/144
2 derivatives checked
- mradermacher/Qwen3-Jan-Nano-128k-6B-Brainstorm20x-i1-GGUF: Quantization
- mradermacher/Qwen3-Jan-Nano-128k-6B-Brainstorm20x-GGUF: Same weights, converted