QinghuiWang

add SKEP

1

add SKEP Paper Interpretation.

### PR types New features ### PR changes Others ### Description Add unified sentiment analysis based on UIE.

Something confused me about Nystromformer

3

### System Info Any GPU Machine with Transformer 4.26.0 ### Who can help? @ArthurZucker @younesbelkada @sgugger @novice03 ### Information - [x] The official example scripts - [ ] My own...

[BUG] torch._C._LinAlgError: linalg.cholesky always raised

5

**Describe the bug** Hi, @PanQiWei @TheBloke Thanks for all contributions in quantization. Recently, I try quantize a llama-style model with 16B params and 32k length，but unfortunately an Exception raised when...

Hi, @byshiue I wanna to quantize a llama model with long sequence 120K+， but an OOM Error raised. So I hope to solve OOM problem with multi gpus when quantizing...

[Quenstion] TensorRT-LLM cannot do inference when sequence length > 200K?

2

Hi, @byshiue @QiJune I build a llama model with long sequence (200K tokens+), but I encouter the problems as follows: - when doing inference with sequence 200K， it raise an...

Caption数量和图片数量不匹配

1

你好，感谢贡献TaiSu数据集。 在数据下载目录中，filtered_caption.txt中caption的数量大约是14M，共计有37个压缩文件，看了下文件中都是图片，图片数量远远大于caption数量。 这个是为什么呢？

hi, @ctlllll I try to use medusa on llama model，and do some medusa head experiments. when base_model_config. medusa_num_heads in from_pretrained(medusa_model.py) is set to be 2 or 3， an error will...

Encounter CUDA error when increasing the length of input_ids

2

### System Info GPU: A800 GPU memory: 80G TensorRT-LLM: 0.8.0 CUDA: 12.1 OS: unbuntu ### Who can help? @byshiue @kaiyux ### Information - [ ] The official example scripts -...