FractalAIResearch/Fathom-Search-4B
Full fine-tune vs Qwen/Qwen3-4B. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 105/145, MLP 108/108, Attention 144/144
4 derivatives checked
- mradermacher/Fathom-Search-4B-GGUF: Same weights, converted
- mradermacher/Fathom-Search-4B-i1-GGUF: Quantization
- prithivMLmods/Fathom-4B-AIO-GGUF: Partial fine-tune
- ZeroXClem/Qwen3-4B-CrystalSonic: Full fine-tune