Mawdistical-S1/Infracelestial-7B

Full fine-tune vs Mawdistical/Kuwutu-7B. Fine-tune: 100% of the weights changed, across 36 of 36 layers.

Tensors changed: Embeddings 1/1, Norms 0/73, MLP 108/108, Attention 144/252, lm_head 1/1

3 derivatives checked

• bartowski/Mawdistical-S1_Infracelestial-7B-GGUF: Same weights, converted
• mradermacher/Infracelestial-7B-i1-GGUF: Quantization
• mradermacher/Infracelestial-7B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse