AnswerDotAI fsdp_qlora issues

AnswerDotAI / fsdp_qlora

Training LLMs with QLoRA + FSDP

Apache License 2.0

1.38k stars 183 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Fix: RuntimeError, Error(s) in loading state_dict for PeftModelForCau…

#71 chwenjun225 opened 1 week ago
0
`Converting the State Dict.ipynb` - Runtine error because Unexpected keys

#70 chwenjun225 closed 1 week ago
1
How to fine-tune a Vision Language Model (VLM)?

#69 asmith26 opened 1 week ago
0
DoRA training not taking dropout or alpha into account

#68 BenjaminBossan opened 3 weeks ago
0
[FEATURE] Profiling Improvements

#67 jeromeku closed 2 months ago
7
Add profiling to train.py

#66 austinvhuang closed 3 months ago
1
Create benchmarks_03_2024.md

#65 johnowhitaker closed 4 months ago
0
Add n_bits param to pass to hqq

#64 UmerHA closed 4 months ago
1
train.py

#63 mylesgoose opened 4 months ago
1
fix multiprocess issue (RuntimeError An attempt has been made to start a new process before the current process has finished its bootstrapping phase)

#62 geronimi73 opened 4 months ago
0
ValueError report

#61 mxjmtxrm opened 4 months ago
0
Request for Scripts to Merge QDoRA Adapters with Base Model for vLLM Inference

#60 iseesaw opened 4 months ago
4
Question about GPU memory usage.

#59 mxjmtxrm opened 4 months ago
0
DeepSeek VL support

#58 SinanAkkoyun opened 4 months ago
0
How does one load and do inference on fine-tuned LLama 3 using bnb_dora train script?

#57 pe-hy opened 4 months ago
0
BOFT support?

#56 Xynonners opened 4 months ago
0
Can i use this script to pre-train models?

#55 brunocruzfranchi opened 4 months ago
0
Issues with LLaMA-3-70B

#54 catid closed 4 months ago
1
llama3?

#53 lianuo opened 4 months ago
0
Llama pro

#52 KeremTurgutlu closed 4 months ago
0
DoRA

#51 KeremTurgutlu closed 4 months ago
0
Results after running

#50 hsb1995 opened 4 months ago
0
What if I have three graphics cards?

#49 lianuo opened 4 months ago
1
How to load the saved model?

#48 bilalghanem opened 5 months ago
0
process 0 terminated with signal SIGKILL

#47 hsb1995 opened 5 months ago
4
how to inference using 70b? or we need to implement it with the same way to train it by ourself?

#46 yaohwang opened 5 months ago
1
nan when the input length is large

#45 bilalghanem opened 5 months ago
5
Why is o_proj not targetted?

#44 GRcharles opened 5 months ago
0
Question about adding / training Mixtral

#43 chrismrutherford opened 5 months ago
1
Q on comparison with SFTTrainer

#42 RonanKMcGovern opened 5 months ago
0
Dual GPU training instantly powers off my desktop

#41 rationalism closed 4 months ago
6
/opt/conda/conda-bld/pytorch_1708025847130/work/aten/src/ATen/native/cuda/Loss.cu:250: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [14,0,0] Assertion `t >= 0 && t < n_classes` failed.

#40 yaohwang opened 5 months ago
0
print reserved memory, log allocated & reserved

#39 warner-benjamin closed 5 months ago
0
Bigger context size?

#38 LoganALJones opened 5 months ago
0
train.py script crashes when using HQQ

#37 rationalism opened 5 months ago
3
(minor) add type hints to train.py

#36 Liberatedwinner closed 5 months ago
0
Add dataset_samples argument for alpaca_sample and dummy datasets

#35 warner-benjamin closed 5 months ago
0
Torch Compile?

#34 jzhang38 opened 5 months ago
0
Support torch model bin format

#33 okdshin opened 5 months ago
0
Example with AMD ROCm/HIP

#32 ehartford closed 5 months ago
4
Adding NVMe SSDs to Enable and Accelerate 100B Model Fine-tuning on a Single GPU

#31 Iron-Bound opened 5 months ago
0
move peft imports to avoid RuntimeError: An attempt has been made to start a new process before the current process has finished its bootstrapping phase

#30 geronimi73 closed 5 months ago
17
Fine tuning only runs on CPU

#29 diabeticpilot opened 6 months ago
4
RuntimeError: An attempt has been made to start a new process before the current process has finished its bootstrapping phase

#28 geronimi73 closed 5 months ago
3
Update 00-profile_lora_qlora.ipynb

#27 eltociear opened 6 months ago
0
bugs for fine-tune fsdp multinode

#26 batman-do opened 6 months ago
1
Running into CUDA out of memory with hqq_lora

#25 zabirauf closed 5 months ago
3
ProcessExitedException: process 0 (2x 4090)

#24 Pugio opened 6 months ago
39
Update README.md

#23 Xorlent closed 6 months ago
1
NCCL issue training with two GPUs

#22 deepankarsharma opened 6 months ago
2