mobilint/Qwen2.5-0.5B-Instruct-Batch16
Full fine-tune (fewer layers) vs Qwen/Qwen2.5-0.5B-Instruct. mobilint/Qwen2.5-0.5B-Instruct-Batch16 matches part of Qwen/Qwen2.5-0.5B-Instruct (1 of its 290 tensors; the rest of Qwen/Qwen2.5-0.5B-Instruct has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 0 of 0 layers.
Tensors changed: Embeddings 1/1
2 other repos hold byte-for-byte identical weight files.