attashe/q-2.5-deepseek-r1-veltha-v0.3

Full fine-tune vs djuna/Q2.5-Veltha-14B. Fine-tune: 100% of the weights changed, across 48 of 48 layers.

Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 97/97, MLP 144/144, Attention 336/336

1 derivatives checked

• mradermacher/q-2.5-deepseek-r1-veltha-v0.3-GGUF: Quantization

On the Hub · Downloads on Model Pulse