-
I have encountered the problem mentioned in the title. Could someone help me understand what is going on and how to resolve it?
Any assistance would be greatly appreciated.
-
Hi, How to check the confidence score or probability score of the generated text? Throwing me error when I try to find the logits
-
Across different LMMs the max new token is different .
I believe we should have a consistent MAX_NEW_TOKENS across the project, set to 512 or 1024
If it makes sense, I can create a PR to modify al…
-
### System Info / 系統信息
tokenizer = LlamaTokenizer.from_pretrained('lmsys/vicuna-7b-v1.5')
model = AutoModelForCausalLM.from_pretrained(
'THUDM/cogvlm-chat-hf',
torch_dtype=torch.bfloat16,
…
-
Thansk for the great work. It's very intersting.
Could you please provide some examples on running the evaluation? The code in VLMEvalkit folder seems incomplete.
And are you considering adding t…
-
```
Traceback (most recent call last):
File "/ssd/ylying/CogVLM/basic_demo/infer_dataset.py", line 164, in
main()
File "/ssd/ylying/CogVLM/basic_demo/infer_dataset.py", line 36, in main
…
-
Any plan to support the latest Qwen2-VL model evaluation?
-
请问CogVLM大模型断点继续训练如何设置配置
![image](https://github.com/THUDM/CogVLM/assets/37207093/37c0702f-8e71-4263…
-
感谢开源！
请问首页README说的要发布基于GLM的中文CogVLM，是这个模型吗？ https://chatglm.cn/ 。体验下来感觉中文效果跟英文还是有明显差距。加油！
什么时候能开源呢？
-
I am able to use V1 perfectly fine on Windows 10
But V2 depends on Triton
Even though I have installed pre-compiled Triton package still i am getting below error
```
Error: unsupported opera…