Indexnusrefather/Pallete-2.6B-RP-Reasoning-2609-v1b
Partial fine-tune vs LiquidAI/LFM2.5-2.6B. Partial change: 22% of the weights changed (attention, mlp, other), in 26 of 30 layers.
Tensors changed: Embeddings 0/2, other 18/66, MLP 20/90, Norms 0/76, Attention 30/32
2 derivatives checked
- mradermacher/Pallete-2.6B-RP-Reasoning-2609-v1b-i1-GGUF: Quantization
- mradermacher/Pallete-2.6B-RP-Reasoning-2609-v1b-GGUF: Same weights, converted