uw-math-ai/MathLeap-Octen-8B

Full fine-tune vs Octen/Octen-Embedding-8B. Fine-tune: 92% of the weights changed, across 36 of 36 layers.

Tensors changed: Embeddings 0/1, Norms 95/145, MLP 108/108, Attention 144/144

1 derivatives checked

• mradermacher/MathLeap-Octen-8B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse