convaiinnovations/ECG-Instruct-Llama-3.2-11B-Vision

Partial fine-tune vs unsloth/Llama-3.2-11B-Vision-Instruct. Partial change: 89% of the weights changed (attention, mlp, other), in 40 of 40 layers.

Tensors changed: Embeddings 0/10, Norms 0/261, MLP 200/288, Attention 320/344, lm_head 0/1, other 1/2

1 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse