suayptalha/Rombos-2.5-T.E-8.1

Full fine-tune vs Cran-May/T.E-8.1. Fine-tune: 97% of the weights changed, across 28 of 28 layers.

Tensors changed: Embeddings 1/1, Norms 19/57, MLP 81/84, Attention 131/196, lm_head 1/1

2 derivatives checked

• mradermacher/Rombos-2.5-T.E-8.1-i1-GGUF: Quantization
• mradermacher/Rombos-2.5-T.E-8.1-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse