AnandSingh/hunyuanocr-mlx
Full fine-tune vs tencent/HunyuanOCR. Fine-tune: 100% of the weights changed, across 27 of 27 layers.
Tensors changed: lm_head 1/1, Embeddings 3/3, Norms 205/205, MLP 182/182, Attention 312/312, other 6/6
Full fine-tune vs tencent/HunyuanOCR. Fine-tune: 100% of the weights changed, across 27 of 27 layers.
Tensors changed: lm_head 1/1, Embeddings 3/3, Norms 205/205, MLP 182/182, Attention 312/312, other 6/6