qgallouedec/SmolLM2-360M-Rickified

Full fine-tune vs HuggingFaceTB/SmolLM2-360M. 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

2 derivatives checked

• mradermacher/SmolLM2-360M-Rickified-GGUF: Same weights, converted
• chrishoertnagl/SmolLM2-360M-Rickified-GRPO: Full fine-tune

On the Hub · Downloads on Model Pulse