John

IlikeLLM's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in unsloth/Qwen3-VL-8B-Thinking-FP8 9 months ago

Why is the pad token of all QWEN VL models in Unsloth "<|vision_pad|>", while QWEN officially uses "pad_token": "<|endoftext|>"

1

#1 opened 9 months ago by

IlikeLLM

Why is the pad token of all QWEN VL models in Unsloth "<|vision_pad|>", while QWEN officially uses "pad_token": "<|endoftext|>"

1

#1 opened 9 months ago by

IlikeLLM