CloudGoat/Mephisto-4B-0725

Full fine-tune (fewer layers) vs InternScience/Agents-A1-4B. CloudGoat/Mephisto-4B-0725 matches part of InternScience/Agents-A1-4B (426 of its 723 tensors; the rest of InternScience/Agents-A1-4B has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Attention 176/224, MLP 96/96, Embeddings 1/1, Norms 42/105

On the Hub · Downloads on Model Pulse