vllm-sr/Vela-1.0-Encoder-307M-Hazard
Full fine-tune vs vllm-sr/Vela-1.0-Encoder-307M. Fine-tune: 100% of the weights changed, across 22 of 22 layers.
Tensors changed: other 1/1, Norms 45/45, Embeddings 2/2, Attention 44/44, MLP 44/44
Full fine-tune vs vllm-sr/Vela-1.0-Encoder-307M. Fine-tune: 100% of the weights changed, across 22 of 22 layers.
Tensors changed: other 1/1, Norms 45/45, Embeddings 2/2, Attention 44/44, MLP 44/44