OpenLLM-Ro/RoGemma-7b-Instruct

Full fine-tune vs google/gemma-7b. Fine-tune: 100% of the weights changed, across 28 of 28 layers.

Tensors changed: Embeddings 1/1, Norms 53/57, MLP 84/84, Attention 112/112

2 derivatives checked

• legraphista/RoGemma-7b-Instruct-IMat-GGUF: Full fine-tune
• mradermacher/RoGemma-7b-Instruct-GGUF: Full fine-tune

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

On the Hub · Downloads on Model Pulse