issues
search
Lightning-AI
/
litgpt
Pretrain, finetune, deploy 20+ LLMs on your own data. Uses state-of-the-art techniques: flash attention, FSDP, 4-bit, LoRA, and more.
https://lightning.ai
Apache License 2.0
6.85k
stars
726
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Adds streaming option to generate
#1424
rasbt
opened
2 days ago
4
Continually pretrained Llama2-7B-hf model inference is not working on 16GB GPU machine
#1423
karkeranikitha
opened
2 days ago
4
Resolve output characters garbled
#1422
fireyanci
opened
3 days ago
1
WIP: TensorParallel with new strategy
#1421
awaelchli
opened
4 days ago
0
Stream option
#1420
rasbt
opened
4 days ago
3
Python API
#1419
rasbt
opened
4 days ago
0
how to pretrain llama2?
#1418
wen020
opened
4 days ago
1
Lora recipes use lots of memory because of not wrapping parameters with gradient in separate FSDP unit
#1417
RuABraun
closed
5 days ago
2
prompt_style
#1416
fireyanci
opened
5 days ago
1
Thunder FSDP as a regular transform
#1415
carmocca
opened
5 days ago
0
Support `no_sync` with Thunder FSDP
#1414
carmocca
closed
5 days ago
0
Continue pre-training got RuntimeError: Failed processing /tmp/data
#1413
BestJiayi
opened
5 days ago
4
'Phi-3-mini-4k-instruct' is not a supported config name
#1412
georgehu0815
opened
6 days ago
1
support for qwen2 and baichuan
#1411
bestpredicts
opened
1 week ago
0
Minimal Python example
#1410
rasbt
closed
5 days ago
0
OptimizerArgs
#1409
rasbt
opened
1 week ago
4
Redundancy?
#1408
rasbt
closed
1 week ago
2
Explain dataset options
#1407
rasbt
closed
1 week ago
0
Add prompt style mapping for llama3
#1406
davmacario
opened
1 week ago
0
Simillar -> Similar
#1405
rasbt
closed
1 week ago
0
LitData: set `iterate_over_all` to False for `CombinedStreamingDataset`
#1404
Andrei-Aksionov
closed
1 week ago
0
Streamline LitGPT API
#1403
rasbt
opened
1 week ago
7
Pretraining example from readme fails in Colab
#1402
AndisDraguns
opened
1 week ago
3
Remove old and unused LLMs
#1401
rasbt
opened
1 week ago
0
Set litdata < 0.2.6
#1400
carmocca
closed
1 week ago
0
test_tinyllama issue with LitData and `iterate_over_all`
#1399
Andrei-Aksionov
closed
1 week ago
2
LoRA matrices dropout
#1398
belerico
opened
1 week ago
0
Remove per-file CLIs
#1397
carmocca
closed
1 week ago
1
Fix issues with LitGPT Tokenizer (SentencePiece and HF Tokenizers)
#1396
Andrei-Aksionov
opened
1 week ago
0
LoRA test: check that all the tensors are materialized.
#1395
Andrei-Aksionov
closed
1 week ago
0
how to solve this debug
#1394
Learneducn
opened
1 week ago
0
LR scheduler can result in a division by 0
#1393
carmocca
opened
1 week ago
0
Address frozen parameter warning with FSDP on nightly torch
#1392
carmocca
opened
1 week ago
1
Unskip Thunder FSDP test
#1391
carmocca
closed
1 week ago
0
Remove duplicate MistralForCausalLM test
#1390
carmocca
closed
1 week ago
0
Fix top-p test
#1389
carmocca
closed
1 week ago
0
Customizable loss function & inference step?
#1388
Boltzmachine
closed
1 week ago
2
ValueError: Cannot attend to 3063, block size is only 2048
#1387
Gooooooogo
closed
1 week ago
1
Will CycleIterator forward to dataset on resume for pretrain?
#1386
calvintwr
opened
1 week ago
1
LoRA multi-GPU no longer works if applying LoRA selectively
#1385
awaelchli
closed
1 week ago
3
fabric.print only works on sys.stderr, does not print inference result
#1384
lastmjs
opened
2 weeks ago
1
How to use custom dataset for evaluate?
#1383
Gooooooogo
opened
2 weeks ago
0
Remove out directory to gitignore
#1382
usmanxia
closed
2 weeks ago
1
Add release workflow
#1381
rasbt
closed
1 week ago
5
Readme.md - Instruct how to get HF_TOKEN
#1380
natanloterio
closed
2 weeks ago
2
How to specify which GPU to use?
#1379
Gooooooogo
closed
2 weeks ago
2
Cannot copy out of meta tensor; no data!
#1378
Gooooooogo
closed
1 week ago
2
After some iteration in pretraining a LLM, IndexError is raised related to dataset chunking
#1377
MusulmonLolayev
opened
2 weeks ago
0
Update LoRA test
#1376
awaelchli
closed
2 weeks ago
0
Update Lightning version
#1375
awaelchli
closed
2 weeks ago
0
Next