coder3101/LFM2.5-8B-A1B-heretic
Full fine-tune vs LiquidAI/LFM2.5-8B-A1B-Base. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: MLP 2162/2162, Embeddings 2/2, other 54/54, Norms 60/60, Attention 24/24
5 derivatives checked
- mradermacher/LFM2.5-8B-A1B-heretic-i1-GGUF: Quantization
- mradermacher/LFM2.5-8B-A1B-heretic-GGUF: Same weights, converted
- FadedRedStar/LFM2.5-8B-A1B-heretic-GGUF: Quantization
- FadedRedStar/LFM2.5-8B-A1B-heretic-imatrix-GGUF: Quantization
- texassred/LFM2.5-8B-A1B-heretic-16e-GGUF: Same weights, repacked (fewer layers)