issues
search
modal-labs
/
llm-finetuning
Guide for fine-tuning Llama/Mistral/CodeLlama models and more
MIT License
537
stars
83
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Fix broken README links
#79
kramstrom
closed
3 months ago
1
Could vllm batch inference be set up?
#78
tonghuikang
closed
3 months ago
1
How do you call an inference from Python?
#77
tonghuikang
closed
3 months ago
1
fix small nits
#76
charlesfrye
closed
3 months ago
0
Updates for Llama 3.1 compatibility, import error workarounds & more
#75
AgenP
opened
3 months ago
0
URGENT: Unable to Train using Modal
#74
vprateek1729
closed
4 months ago
1
Remove notebook
#73
advay-modal
closed
4 months ago
2
Resolve CI-CD failures from volumes without weights
#72
charlesfrye
opened
4 months ago
0
drop deep speed, use bnb 8bit adam optimizer
#71
charlesfrye
closed
4 months ago
0
faster CI
#70
charlesfrye
closed
4 months ago
1
updates vllm to newest version
#69
charlesfrye
closed
4 months ago
0
silence unnecessary warnings
#68
charlesfrye
closed
4 months ago
0
updates axo, vllm
#67
charlesfrye
closed
4 months ago
0
Suggestion: merge-only option
#66
psimm
opened
5 months ago
0
Pin numpy<2 in vLLM image
#65
mwaskom
closed
5 months ago
0
Following the quickstart example results in an error
#64
holma91
closed
5 months ago
4
DeepSpeed documentation corrections
#63
lhl
opened
5 months ago
0
handle tokenization more correctly, clean up CI
#62
charlesfrye
closed
6 months ago
1
adds wandb to CI, fixes seed
#61
charlesfrye
closed
6 months ago
1
better handling of pre-processing and training++
#60
charlesfrye
closed
6 months ago
1
Better fix for wandb secret
#59
aksh-at
closed
6 months ago
0
Fix optional W&B secret
#58
aksh-at
closed
6 months ago
0
better handling of pre-processing and training
#57
winglian
closed
6 months ago
2
use flag to disable wandb from CLI
#56
charlesfrye
closed
6 months ago
0
Fix typos in README.md
#55
franklynwang
closed
7 months ago
0
Update fine-tuning repo
#54
hamelsmu
closed
6 months ago
1
Do not hardcode LoRA merging
#53
hamelsmu
closed
7 months ago
1
stub -> app
#52
erikbern
closed
7 months ago
0
Audience For This Repo
#51
hamelsmu
opened
7 months ago
3
Do not hardcode LORA merging
#50
hamelsmu
closed
7 months ago
0
fix nits in docs
#49
hamelsmu
closed
7 months ago
1
Remove GUI and Update Docs
#48
hamelsmu
closed
7 months ago
10
remove GUI
#47
hamelsmu
closed
7 months ago
0
Pin wandb version in llm-finetuning repo
#46
yirenlu92
closed
7 months ago
5
Proposal: Get Rid of Gradio Interface
#45
hamelsmu
closed
7 months ago
1
TODO: hamel to explore GUIs
#44
hamelsmu
closed
7 months ago
1
update docs
#43
hamelsmu
closed
7 months ago
4
Pin typer version to avoid error
#42
yirenlu92
closed
7 months ago
0
getting TypeError while launching the training job
#41
arpitingle
closed
7 months ago
1
Better detection of last model
#40
erikbern
closed
8 months ago
0
Add lightweight Pythia config and a few small tweaks
#39
mwaskom
closed
8 months ago
1
Add web endpoint to the inference server
#38
erikbern
closed
8 months ago
1
Use output_dir as specified in config
#37
mwaskom
closed
8 months ago
1
Make the post-training steps a bit more clear
#36
erikbern
closed
8 months ago
0
Fix deprecation warning with Volume constructor
#35
erikbern
closed
8 months ago
0
try fix ci
#34
gongy
closed
9 months ago
2
add inference to ci
#33
gongy
closed
9 months ago
0
Update codebase to work better with Mixtral out of the box
#32
gongy
closed
9 months ago
0
Working Mixtral support
#31
gongy
closed
9 months ago
0
Update GUI code to reflect recent config/data changes
#30
mwaskom
closed
9 months ago
1
Next