ngqtrung/Video-HopChain-8B-Standard-RL

Full fine-tune vs Qwen/Qwen3-VL-8B-Instruct. Fine-tune: 100% of the weights changed, across 36 of 36 layers.

Tensors changed: Embeddings 4/4, Norms 172/261, MLP 219/232, Attention 252/252, lm_head 1/1

1 derivatives checked

• prithivMLmods/Video-HopChain-8B-Standard-RL-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse