TheDrummer/Skyfall-39B-v1

Full fine-tune vs TheDrummer/MS-Interleaved-Upscale-39B. Fine-tune: 99% of the weights changed, across 99 of 99 layers.

Tensors changed: lm_head 1/1, Embeddings 0/1, Norms 133/199, MLP 297/297, Attention 396/396

2 derivatives checked

• bartowski/Skyfall-39B-v1-GGUF: Same weights, converted
• mradermacher/Skyfall-39B-v1-i1-GGUF: Quantization

On the Hub · Downloads on Model Pulse