T145/KRONOS-8B-V9

Partial fine-tune vs T145/KRONOS-8B-V1-P1. Partial change: 1% of the weights changed (attention, norms), in 11 of 32 layers.

Tensors changed: Embeddings 0/1, Norms 1/65, MLP 0/96, Attention 11/128, lm_head 0/1

2 derivatives checked

• mradermacher/KRONOS-8B-V9-i1-GGUF: Quantization
• mradermacher/KRONOS-8B-V9-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse