-
### 🐛 Describe the bug
''' checkpoint_path = './llama_relevance_results'
training_args = transformers.TrainingArguments(
#remove_unused_columns=False, # Whether or not to automatically r…
-
Hi!
I recently learned (thanks @skochinsky ) that on IDA, typing D or A on a global variable to create a BYTE/WORD/DWORD/QWORD or a string, does not actually store type information. So changing a glo…
-
They don't support module hooks of global modules, such as OnModuleUnloading.
For python, except adding them to module.h, functions.in and znc.py, need to fix codegen.pl to check if GetUser() returned…
-
### System Info
```
Ubuntu 20.04
Python 3.10.14
torch 2.3.0
transformers 4.42.3
bitsandbytes 0.42.0
CUDA Version: 12.4
GPU 3090
torch.cuda.is_avai…
-
I am unable to use the sample Mamba2 code. Even with following simple code, it fails to do forward pass.
```
import torch
from mamba_ssm import Mamba2
batch, length, dim = 2, 64, 256
x = torch.…
-
Run DDP with a shared buffer (different TorchDynamo `Source`):
Repro Script
```
"""
torchrun --standalone --nproc_per_node=1 test/dup_repro.py
TORCH_LOGS=aot,dynamo torchrun --standalone --…
-
### Checklist
- [X] I have searched related issues but cannot get the expected help.
- [X] 2. I have read the [FAQ documentation](https://github.com/open-mmlab/mmdeploy/tree/main/docs/en/faq.md) b…
-
I played around with the library a bit and found that I can customize different behaviors by subclassing the sqlalchemy-file classes. I can do it without modifying the library code at all, the only th…
-
Hello, can you provide code to run inference of this repo? thank you so much.
-
from trl import SFTTrainer
from transformers import TrainingArguments
from unsloth import is_bfloat16_supported
trainer = SFTTrainer(
model = model,
tokenizer = tokenizer,
train_da…