issues
search
Lightning-AI
/
lit-llama
Implementation of the LLaMA language model based on nanoGPT. Supports flash attention, Int8 and GPTQ 4bit quantization, LoRA and LLaMA-Adapter fine-tuning, pre-training. Apache 2.0-licensed.
Apache License 2.0
5.97k
stars
518
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
readme: add maintenance warning
#491
Borda
closed
3 weeks ago
0
add special token
#490
shunlinlu
opened
3 months ago
0
数据集与训练方法的相关问题
#489
apscat
closed
4 months ago
0
Where is tokenizer.model? tokenizer path
#488
andreamigliorati
opened
5 months ago
1
Using llama3 through lit lama
#487
fireyanci
opened
5 months ago
5
How to convert lit-llama pretrained model to HF format?
#486
karkeranikitha
opened
5 months ago
1
why cannot the generate function be used twice
#485
WyGongya
opened
6 months ago
0
Converting from lit-llama to HF checkpoint?
#484
jacqueline-he
opened
6 months ago
0
it seems that hash of traindata is lost, so it's impossible to continue finetune after stop
#483
drazdra
opened
6 months ago
0
`PackedDatasetBuilder` does not separate with `sep_token`
#482
calvintwr
opened
6 months ago
0
OSError: Not found: "checkpoints/lit-llama/tokenizer.model": No such file or directory Error #2
#481
anirudhitagi
closed
7 months ago
4
TPU Training
#480
kathir-ks
closed
3 months ago
0
Issue with Rotary Embedding Initialization when the number of devices is > 1
#479
diegodoimo
closed
8 months ago
0
Add license at the top of files
#478
lantiga
closed
1 month ago
1
Beam search generation
#477
hellowoe23
opened
9 months ago
0
Fix: Variable name
#476
dhdbsrlw
opened
9 months ago
0
Error: git submodule update --init --recursive -q did not run successfully
#475
seupedro
opened
9 months ago
1
Ban some tokens
#474
AnnaKholkina
opened
10 months ago
0
Can I use Lightning fabirc to pre train llama2 on v100?
#473
JerryDaHeLian
opened
10 months ago
0
RuntimeError: Expected x1.dtype() == cos.dtype() to be true, but got false. (Could this error message be improved? If so, please report an enhancement request to PyTorch.)
#472
JerryDaHeLian
opened
10 months ago
1
RuntimeError: cutlassF: no kernel found to launch!
#471
xvanQ
closed
10 months ago
1
How to quantize LLama in fine-tuning ?
#470
sfarzi
opened
10 months ago
0
How to convert hf weight of 70b to lit-lamma weights?
#469
sfarzi
opened
10 months ago
0
Full fine-tuning on Alpaca dataset with 4 L40s GPUs fails 8 hours into the training job with index_copy_
#468
cabal-daniel
closed
10 months ago
2
RuntimeError: probability tensor contains either `inf`, `nan` or element < 0
#467
AI-Zebra
opened
11 months ago
0
Why is LLaMA response to queries in the conversation so wrong?
#466
Harsh-raj
opened
11 months ago
0
Running into StopIteration with single node multi GPU pretraining against the redpajama sample
#465
cabal-daniel
closed
11 months ago
6
How to do conversation with fine tuned model?
#464
Harsh-raj
closed
11 months ago
0
[question] nan loss value and run time error
#463
nevermet
opened
11 months ago
0
[question] assert lora_path.is_file() error
#462
nevermet
opened
11 months ago
0
Question about 'validating...' from lora.py
#461
nevermet
opened
11 months ago
0
[question] error message while finetuning
#460
nevermet
opened
11 months ago
2
Adapter finetuning do not run on two cards (A100 40G)
#459
wasifferoze
opened
1 year ago
0
Mistral Model
#458
PierreColombo
opened
1 year ago
1
When I finetuned the model, an error occurred during the decoding process: IndexError: Out of range: piece id is out of range.
#457
HypherX
opened
1 year ago
0
ci: labels PR conflicts
#456
Borda
closed
11 months ago
0
adaptor_v2.py should support multi gpu mode.
#455
MartinForReal
closed
9 months ago
0
adding check for valid GHA & broken links
#454
Borda
closed
1 year ago
0
fix requirements + aggregate setup & req. & config in pyproject
#453
Borda
closed
1 month ago
0
Looking for LLaMA 2?
#452
carmocca
opened
1 year ago
3
Only add a linear layer to LLaMA without any computation degrade the performance
#451
YUCHEN005
opened
1 year ago
0
How can I do to inferece with different promts in Juypter Notebook, just load the model and tokenizer once?
#450
Vinter8848
closed
1 year ago
2
(documentation) How do I know if generate.py is running on GPU / GPU configuration
#449
maathieu
opened
1 year ago
1
(documentation) error on readme.md about (Facebook's) LLama's license
#448
maathieu
closed
1 year ago
1
ValueError: Precision 'bf16-true' is invalid
#447
AlphaGoMK
closed
1 year ago
1
multi gpus for full finetune
#446
qiqiApink
opened
1 year ago
3
TypeError: super(type, obj): obj must be an instance or subtype of type
#445
Vinter8848
opened
1 year ago
3
Check torch and lightning versions before running models
#444
rasbt
opened
1 year ago
0
No response after training an epoch
#443
Dylandtt
opened
1 year ago
2
inference finetuned model using LoRa in Huggingface format
#442
LamOne1
closed
1 year ago
7
Next