Magpie-Align/MagpieLM-4B-Chat-v0.1
Full fine-tune vs Magpie-Align/MagpieLM-4B-SFT-v0.1. Fine-tune: 91% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 13/65, MLP 96/96, Attention 128/128, lm_head 1/1
5 derivatives checked
- mradermacher/MagpieLM-4B-Chat-v0.1-i1-GGUF: Quantization
- Felladrin/gguf-Q5_K_M-MagpieLM-4B-Chat-v0.1: Quantization
- mradermacher/MagpieLM-4B-Chat-v0.1-GGUF: Same weights, converted
- bunnycore/LLama-3.1-4B-TitanFusion: Full fine-tune
- win10/MagpieLM-8B: Partial fine-tune (fewer layers)