-
The last meaningful output:
```
ram used: 5.67 GB, tok_embeddings.weight : 99%|▉| 290/292 [00:25
-
I try to load loras with
pipeline.pipe.load_lora_weights("/kaggle/input/lorass/acuarelac1400.safetensors")
I don't know if it is the correct way, it would be helpful if you told me how to load lor…
-
I encountered an issue when attempting to trace a CrossEncoder model using torch.jit.trace. The error occurs during the tracing process when calling the forward method. Below is a minimal reproducible…
-
### 🐛 Describe the bug
I'm compiling a graph multiple times using inductor. I find it inplace modify the graph, and one of the graph's output changes from tensor to list of tensor.
example code:
…
-
**Describe the bug**
I am using the `train_gpt3_175b_distributed.sh` script to launch training on a single node with 4 A100 80GB GPUs. Training goes well if I use tensor parallel or pipeline parallel,…
-
debug inference过程发现视觉模块进行图片编码就需要消耗3GB以上的显存，调试发现主要多层self-attention模型消耗的，实际我单独将视觉模型保存下来，发现仅360M左右大小
@Ucas-HaoranWei 请帮忙看看其中是为啥？
-
## 🚀 Feature
Improved Transformer and MultiHeadAttention design
## Motivation
Current Transformer and MultiHeadAttention design is not a very `PyTorch`-ish. The `in_proj` is not even a `nn.Mo…
-
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
Cell In[16], line 3
1 generate_text = I…
-
### Feature
Adjust the positional encoding when compressing the cache to improve output quality for long sequences. Based on our experiments, this adjustment significantly reduces gibberish output and…
-
**Describe the bug**
I try to use LLaVA example and faced to key mismatch error. I am on latest commit in main branch. (094d66b)
[rank0]: RuntimeError: Error(s) in loading state_dict for LLaVAMode…