virtuous7373/Gemma-4-Harmonia-31B

Full fine-tune vs ConicCat/Gemma4-GarnetV2-31B. Fine-tune: 100% of the weights changed, across 60 of 60 layers.

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

3 derivatives checked

• mradermacher/Gemma-4-Harmonia-31B-i1-GGUF: Quantization (fewer layers)
• mradermacher/Gemma-4-Harmonia-31B-GGUF: Quantization (fewer layers)
• llmfan46/Gemma-4-Harmonia-31B-uncensored-heretic: Abliterated

On the Hub · Downloads on Model Pulse