allenai/Olmo-3-7B-Instruct-DPO

Full fine-tune vs allenai/Olmo-3-7B-Instruct-SFT. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 1/1, MLP 96/96, Norms 128/129, Attention 128/128, lm_head 1/1

4 derivatives checked

• allenai/Olmo-3-7B-Instruct: Full fine-tune
• Mungert/Olmo-3-7B-Instruct-GGUF: Full fine-tune
• mradermacher/Olmo-3-7B-Instruct-DPO-i1-GGUF: Quantization
• mradermacher/Olmo-3-7B-Instruct-DPO-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse