shlv/AudioJev
Partial fine-tune (fewer layers) vs Qwen/Qwen2.5-Omni-3B. shlv/AudioJev matches part of Qwen/Qwen2.5-Omni-3B (1442 of its 2544 tensors; the rest of Qwen/Qwen2.5-Omni-3B has no counterpart). On the shared part: Partial change: 73% of the weights changed (attention, mlp, norms, other), in 36 of 36 layers.
Tensors changed: other 6/7, MLP 236/432, Norms 180/268, Attention 476/732, Embeddings 0/2, lm_head 0/1