kristijonas/paprika-whisper-lt
Full fine-tune vs svogunas/whisper-large-v3-turbo-lt. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 2/3, Norms 128/156, Attention 280/280, MLP 144/144, other 4/4
1 derivatives checked
- kristijonas/paprika-whisper-lt-v3: Full fine-tune