arcee-ai/Trinity-Mini-Base

Full fine-tune vs arcee-ai/Trinity-Mini-Base-Pre-Anneal. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 1/1, Norms 193/193, MLP 2645/2645, Attention 160/160, lm_head 1/1

4 derivatives checked

• arcee-ai/Trinity-Mini: Full fine-tune
• Mungert/Trinity-Mini-GGUF: Full fine-tune
• mradermacher/Trinity-Mini-Base-i1-GGUF: Quantization
• mradermacher/Trinity-Mini-Base-GGUF: Quantization

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

On the Hub · Downloads on Model Pulse