SultanR/SmolTulu-1.7b-Instruct
Full fine-tune vs HuggingFaceTB/SmolLM2-1.7B. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Embeddings 1/1, Norms 49/49, MLP 72/72, Attention 96/96
2 derivatives checked
- SultanR/SmolTulu-1.7b-Reinforced: Full fine-tune
- SultanR/SmolTulu-1.7b-RM: Full fine-tune