Datadog/Toto-2.0-2.5B-FT
Full fine-tune vs Datadog/Toto-2.0-2.5B. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: other 12/12, Attention 240/288, MLP 96/144
Full fine-tune vs Datadog/Toto-2.0-2.5B. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: other 12/12, Attention 240/288, MLP 96/144