Issues
- 0
No license file
#32 opened by Maniues - 2
I would like a longer text result
#31 opened by r23 - 0
- 42
Finetuning
#19 opened by Stamenov - 4
How to generate or convert vocab.json, merges.txt, and config.json to match huggingface/transformers requirements ?
#29 opened by ycat3 - 5
Silent failure when training on GPU
#14 opened by vilhub - 10
- 2
Pytorch: Speed up get_log_probs function
#5 opened by binhvq - 3
Question about train dataset format
#17 opened by choomz - 1
"state_dict" Mismatch
#20 opened by nitinnairk - 1
Select GPU of choice
#25 opened by Meghana-Meghana - 6
No speed up when using muli-gpu training
#24 opened by zaidalyafeai - 6
Training GPT-2 on very large corpus
#23 opened by simonefrancia - 7
Fail to resume on multiple gpu
#21 opened by knok - 2
- 5
Validation loss not computed
#16 opened by nitinnairk - 2
Unigram algorithm instead of BPE
#15 opened by nitinnairk - 2
Plans for transformer-xl?
#13 opened by gooofy - 2
Plans to add gradient checkpointing?
#11 opened by gooofy - 7
- 7
Training from scratch - how many epochs?
#8 opened by gooofy - 0
Predict with GPU
#7 opened by binhvq - 1
Error on validate, batch is empty
#6 opened by binhvq - 10
Train in large dataset
#3 opened by binhvq - 6