Issues
- 3
checkpoint's size is increasing everytime.
#134 opened by IvoryTower800 - 1
Import Error
#150 opened by mohammad0081 - 3
Mosaic kernels cannot be automatically partitioned. Please wrap the call in a shard_map or xmap
#149 opened by nyl199310 - 5
- 14
Kaggle training examples don't work
#140 opened by jcole75 - 3
Out of memory for serving example
#142 opened by xu3kev - 7
Can't load checkpoints continue training
#148 opened by IvoryTower800 - 6
training does not start using latest easydel
#146 opened by IvoryTower800 - 2
load_in_8bit doesn't work on Kaggle TPU
#143 opened by IvoryTower800 - 4
'LoraWeight' object has no attribute 'tolist'
#145 opened by defdet - 4
Please provide support for LLama3 or provide example on how to serve it using Easydel
#144 opened by jchauhan - 8
How to reduce TPU RAM when finetuning?
#131 opened by IvoryTower800 - 3
- 19
Transformers-like API for inference
#128 opened by Froggy111 - 6
Unable to Load EasyDeL State
#133 opened by w11wo - 2
- 6
a question about how to increase batch size.
#125 opened by IvoryTower800 - 9
- 10
Training with Ring Attention Failed
#120 opened by IvoryTower800 - 1
Docs site is broken https://erfanzar.github.io
#121 opened by nigh8w0lf - 8
Install from git not working
#118 opened by sr5434 - 5
Training in kaggle's TPU is failing
#117 opened by saidineshpola - 5
Error while finetuning Tinyllama on Kaggle TPU
#104 opened by jchauhan - 4
- 6
- 1
- 1
Easydel support on TPU v4.8 - getting exception
#111 opened by jchauhan - 3
Example shown on https://pypi.org/project/EasyDeL/ to finetune tinyllama raise exception on kaggle
#105 opened by jchauhan - 1
- 1
Exception while running any model - einops.EinopsError: Error while processing rearrange-reduction pattern "b (c n) d -> b c n d".
#106 opened by jchauhan - 1
None of the examples scripts works, that used to work earlier. Please test your examples again and update docs
#109 opened by jchauhan - 11
QLoRA Finetune Example
#103 opened by sr5434 - 3
Error while training GPT2 on the kaggle
#98 opened by jchauhan - 8
Step time increasing as training progresses
#90 opened by yhavinga - 3
- 2
Text Generation with Mixtral fails
#91 opened by clintg6 - 3
Error while running a GPT2 model
#87 opened by jchauhan - 2
- 2
Error running remote model that has custom code
#94 opened by jchauhan - 2
ValueError: `params` cannot be accessed from model when the model is created with `_do_init=False`.
#96 opened by jchauhan - 1
Error while training a Phi2 model
#93 opened by jchauhan - 14
AMD Hardware Support
#88 opened by ThePerfectComputer - 4
- 9
Training on TPU Using Flash Attention
#83 opened by IvoryTower800 - 10
Inference on Single Node multiple GPUs
#82 opened by clintg6 - 1
- 1
While training Gpt2 model - Exception - TypeError: in_shardings leaf specifications are expected to be PartitionSpec instances or None, but got *
#89 opened by jchauhan - 2
Error while serving model as per documentation,
#80 opened by jchauhan - 1
Mixtral 8x7B support?
#81 opened by clintg6 - 2
Will this project support lora?
#79 opened by IvoryTower800