Minthy/RouWei-0.7

Full fine-tune vs Minthy/RouWei-0.6. Fine-tune: 94% of the weights changed, across 11 of 12 layers.

Tensors changed: Embeddings 568/590, other 534/662, Attention 700/716, Norms 442/546

2 derivatives checked

• Minthy/RouWei-0.8: Full fine-tune
• Emanon14/NONAMEmix-Vpred_v2: Full fine-tune

On the Hub · Downloads on Model Pulse