HuggingFaceTB/smollm-360M-instruct-add-basics
Full fine-tune vs HuggingFaceTB/SmolLM-360M. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 65/65, MLP 96/96, Attention 128/128
3 derivatives checked
- Felladrin/gguf-smollm-360M-instruct-add-basics: Quantization
- HuggingFaceTB/smollm-360M-instruct-v0.2-Q8_0-GGUF: Quantization
- Felladrin/gguf-Q5_K_M-smollm-360M-instruct-add-basics: Quantization
3 other repos hold byte-for-byte identical weight files.