trashpanda-org/gemma-4-31b-larkspur-v1

Full fine-tune vs google/gemma-4-31B. Fine-tune: 100% of the weights changed, across 60 of 60 layers.

Tensors changed: Embeddings 4/4, Norms 105/523, other 0/62, MLP 261/261, Attention 338/338

1 derivatives checked

• mradermacher/gemma-4-31b-larkspur-v1-GGUF: Quantization (fewer layers)

On the Hub · Downloads on Model Pulse