|
About the 🤗Accelerate category
|
|
1
|
2416
|
February 20, 2022
|
|
Loss spike when resuming from FSDP SHARDED_STATE_DICT checkpoint (possible optimizer-state mismatch)
|
|
1
|
12
|
June 28, 2025
|
|
Problem with full-finetuning on cluster
|
|
1
|
17
|
June 25, 2025
|
|
Transformers Trainer + Accelerate FSDP: How do I load my model from a checkpoint?
|
|
3
|
14296
|
June 22, 2025
|
|
NCCL Timeout Accelerate Load From Checkpoint
|
|
2
|
2350
|
June 20, 2025
|
|
Not seeing memory benefit to accelerate/FSDP2
|
|
3
|
36
|
June 18, 2025
|
|
DistributedSampler with Accelerate
|
|
1
|
18
|
June 10, 2025
|
|
Where can I find the full list of parameters for the Accelerate yaml config?
|
|
3
|
20
|
June 5, 2025
|
|
Synchronizing State, Trainer and Accelerate
|
|
3
|
22
|
May 22, 2025
|
|
[RuntimeError] DPOTrainer - "element 0 of tensors does not require grad and does not have a grad_fn" on 8x A100 GPUs
|
|
1
|
33
|
May 20, 2025
|
|
Reproduce SFTTrainer with Accelerate and Pytorch
|
|
0
|
31
|
May 18, 2025
|
|
11B model gets OOM after using deepspeed zero 3 setting with 8 32G V100
|
|
2
|
1250
|
April 26, 2025
|
|
Multi-gpu inference llama-3.2 vision with QLoRA
|
|
4
|
103
|
April 25, 2025
|
|
How to work with meta tensors?
|
|
1
|
2170
|
April 16, 2025
|
|
BitsandBytes conflict with Accelerate
|
|
6
|
491
|
April 14, 2025
|
|
Issues with Dataset Loading and Checkpoint Saving using FSDP with HuggingFace Trainer on SLURM Multi-Node Setup
|
|
1
|
95
|
April 7, 2025
|
|
Meta device error while instantiating model
|
|
5
|
6925
|
April 1, 2025
|
|
Saving bf16 Model Weights When Using Accelerate+DeepSpeed
|
|
4
|
384
|
March 17, 2025
|
|
Cannot run multi GPU training on SLURM
|
|
1
|
111
|
March 16, 2025
|
|
Fp8 error in accelerate test
|
|
1
|
118
|
March 11, 2025
|
|
Accelerator .prepare() replaces custom DataLoader Sampler
|
|
5
|
1283
|
March 9, 2025
|
|
Using large dataset with accelerate
|
|
0
|
43
|
March 6, 2025
|
|
Accelerator.save_state errors out due to timeout. Unable to increase timeout through kwargs_handlers
|
|
5
|
1307
|
March 3, 2025
|
|
HF accelerate DeepSpeed plugin does not use custom optimizer or scheduler
|
|
2
|
23
|
March 1, 2025
|
|
Bug on multi-gpu trainer with accelerate
|
|
6
|
438
|
February 18, 2025
|
|
Accelerate remain stuck on using GPU 5 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular devic
|
|
1
|
994
|
February 17, 2025
|
|
Errors when using gradient accumulation with FSDP + PEFT LoRA + SFTTrainer
|
|
2
|
1073
|
February 6, 2025
|
|
Save accelerate model
|
|
4
|
714
|
February 5, 2025
|
|
Calling other large models at runtime?
|
|
0
|
7
|
February 3, 2025
|
|
Training using FSDP, qLoRa on multinode
|
|
0
|
59
|
January 29, 2025
|