AXCXEPT/phi-4-deepseek-R1K-RL-EZO
Full fine-tune vs microsoft/phi-4. Fine-tune: 93% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 0/1, Norms 1/81, MLP 80/80, Attention 80/80, lm_head 0/1
4 derivatives checked
- mradermacher/phi-4-deepseek-R1K-RL-EZO-i1-GGUF: Quantization
- DevQuasar-8/AXCXEPT.phi-4-deepseek-R1K-RL-EZO-GGUF: Same weights, converted
- mradermacher/phi-4-deepseek-R1K-RL-EZO-GGUF: Quantization
- mergekit-community/Phi-4-reasoning-Line-14b-karcher: Full fine-tune