yanolja/YanoljaNEXT-EEVE-Rosetta-7B-2602

Full fine-tune vs ByteDance-Seed/Seed-X-PPO-7B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Norms 65/65, MLP 96/96, Attention 128/128

2 derivatives checked

• mradermacher/YanoljaNEXT-EEVE-Rosetta-7B-2602-i1-GGUF: Quantization
• mradermacher/YanoljaNEXT-EEVE-Rosetta-7B-2602-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse