issues
search
replit
/
ReplitLM
Inference code and configs for the ReplitLM model family
https://huggingface.co/replit
Apache License 2.0
913
stars
75
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
switch doesn't work in cpp with some cases
#56
ghost
opened
1 month ago
0
Guess
#55
Nelson19so
opened
1 month ago
2
OpenAi_Image_Generator
#54
sandilyakanishk
closed
1 month ago
0
HTML, CSS, JS (10)
#53
Mekodi5
opened
2 months ago
0
Web minta thr @wahyu
#52
Wahyurdi
opened
3 months ago
0
HTML, CSS, JS (9)
#51
Mekodi5
opened
3 months ago
0
HTML, CSS, JS (9)
#50
Mekodi5
opened
3 months ago
0
HTML, CSS, JS (9)
#49
Mekodi5
opened
3 months ago
0
.html
#48
Mekdelawitameha5
opened
3 months ago
0
Login page
#47
Mekdelawitameha5
opened
3 months ago
0
Model fails with AttributeError: 'ReplitLMTokenizer' object has no attribute 'sp_model'
#46
ShisuiMadara
opened
5 months ago
1
Maachudo
#45
Prajjwalp086
opened
6 months ago
0
HTML, CSS, JS
#44
Mathan-kumar-mk
opened
8 months ago
0
GPU hosted demo link broken in README
#43
JohnAllen
opened
8 months ago
0
Clothes
#42
U-kay345
closed
8 months ago
0
Update README.md
#41
madhavatreplit
closed
9 months ago
0
Create replit7
#40
T7x77
closed
8 months ago
0
Challenge 1.2
#39
lathif11
closed
8 months ago
0
Challenge 1.1
#38
lathif11
closed
8 months ago
0
Challenge 1.1
#37
lathif11
closed
9 months ago
0
Replit
#36
Mohitnakhate
closed
8 months ago
0
Sync
#35
U-kay345
closed
8 months ago
0
Challenge 1.1
#34
RahulPriyas
closed
8 months ago
0
Challenge 1.1
#33
RahulPriyas
closed
8 months ago
0
Does ReplitLM support gradient checkpoints?
#32
sxthunder
opened
10 months ago
1
C
#31
Spectre342
closed
8 months ago
0
Using replit_lm_tokenizer locally
#30
Chorleon
closed
11 months ago
1
FlashAttention2 for ReplitLM
#29
kaushal07wick
opened
11 months ago
0
Expected eval results on Multiple-E?
#28
ywen666
opened
11 months ago
0
Replit code can be enhanced with RLHF (Reinforcement learning from human feedback) ?
#27
PuneethSaiCharanRaghavendra
opened
12 months ago
0
Updating eval.py for updated upstream harness code
#26
madhavatreplit
closed
1 year ago
0
Adding evaluation with BigCode harness and patched script
#25
madhavatreplit
closed
1 year ago
0
Speed up the inference
#24
Symbolk
opened
1 year ago
3
Finetuning fails with `RuntimeError: Please install flash-attn==1.0.3.post0 and triton==2.0.0.dev20221202`
#23
matthiasgeihs
opened
1 year ago
2
How to start the server mode?
#22
luckyfish0826
closed
1 year ago
1
How to train the ReplitLM model
#21
ran337287
opened
1 year ago
1
Update README.md for quantization
#20
madhavatreplit
closed
11 months ago
0
Update modeling_mpt.py
#19
madhavatreplit
closed
1 year ago
0
Fix broken link: https://huggingface.co/datasets/bigcode/the-stack-dedup
#18
mishig25
closed
1 year ago
0
Update README with training and instruct tuning guides
#17
madhavatreplit
closed
1 year ago
0
Update weights link in README.md
#16
madhavatreplit
closed
1 year ago
0
Convert ReplitLM to MPT
#15
madhavatreplit
closed
1 year ago
0
Are there any vim plugins?
#14
gunar
opened
1 year ago
1
errors in generation, TypeError: gelu()
#13
ran337287
closed
1 year ago
2
cuda use and out of memory
#12
titoBouzout
closed
1 year ago
5
Fix typo in attention.py
#10
eltociear
closed
1 year ago
1
Warnings and Errors when generating with the given code
#9
Symbolk
closed
1 year ago
2
`ImportError: This modeling file requires... flash_attn`
#8
llimllib
closed
1 year ago
13
where can it run? hardware specs, performance data
#7
elikoga
closed
1 year ago
3
CC BY-SA-4.0 license
#6
MCKevmeister
closed
1 year ago
1
Next