RedHatAI/gemma-4-31B-it-speculator.dspark

Partial fine-tune vs google/gemma-4-31B-it. Partial change: 63% of the weights changed (attention, mlp, norms), in 5 of 5 layers.

Tensors changed: Embeddings 0/1, Norms 21/21, MLP 15/15, Attention 20/20

2 derivatives checked

• williamliao/dspark_gemma4_31b-it-GGUF: Full fine-tune
• Anbeeld/gemma-4-31B-it-DSpark-GGUF: Partial fine-tune

1 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse