issues
search
NVIDIA
/
NeMo-Aligner
Scalable toolkit for efficient model alignment
Apache License 2.0
413
stars
44
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
sync: main to dev
#224
github-actions[bot]
opened
3 days ago
0
fix broken tutorial link
#223
gshennvm
closed
3 days ago
0
Draft+ for SDXL [draft]
#222
rohitrango
opened
5 days ago
0
SPIN: Added logging for avg generation length and generated responses as wandb table
#221
trias702
closed
5 days ago
0
pull changes from degert/spin-trt-beta
#220
gshennvm
closed
5 days ago
0
critic speedup
#219
gshennvm
opened
6 days ago
0
Rejection sampling clean
#218
abukharin3
opened
6 days ago
0
Results do not reproduce between self-hosted and hosted rewards model.
#217
noamgai21
closed
1 day ago
5
Tutorial / Example for Single Node FP8 Inference?
#216
noamgat
opened
1 week ago
0
add fix for missing extra state in rm
#215
gshennvm
closed
1 week ago
0
add fix for missing extra state in rm
#214
gshennvm
closed
1 week ago
0
make num workers all 0 by default
#213
gshennvm
closed
1 week ago
0
fix readme to include technical report
#212
gshennvm
opened
1 week ago
0
add nemotron 4 in docs
#211
gshennvm
closed
1 week ago
0
Rejection sampling clean
#210
abukharin3
closed
6 days ago
0
Rejection sampling clean
#209
abukharin3
closed
1 week ago
0
Rejection sampling
#208
abukharin3
closed
2 weeks ago
0
Multiple training file support
#207
seanliu96
opened
2 weeks ago
1
Add Rejection Sampling
#206
abukharin3
closed
2 weeks ago
0
sync: main to dev
#205
github-actions[bot]
closed
4 days ago
1
Flattening URLs to adhere to best practices.
#204
chrisalexiuk-nvidia
closed
2 weeks ago
0
Updated URLs
#203
jgerh
closed
2 weeks ago
2
Added support for float values for val_check_interval to SFT
#202
trias702
closed
5 days ago
0
Aligner renaming urls
#201
jgerh
closed
2 weeks ago
0
Communicator hang fix in the actor loop
#200
terrykong
opened
2 weeks ago
0
Ensure critic server does not squeeze out a singleton batch dim
#199
terrykong
closed
2 weeks ago
0
Ensure critic server does not squeeze out a singleton batch dim
#198
terrykong
closed
2 weeks ago
0
sync: main to dev
#197
github-actions[bot]
closed
2 weeks ago
0
Small improvements in RM docs
#196
terrykong
closed
2 weeks ago
0
Implement the reward aware preference optimization algorithms.
#195
shengyangs
opened
2 weeks ago
1
Geshen/trt llm to main
#194
gshennvm
opened
3 weeks ago
0
sync: main to dev
#193
github-actions[bot]
closed
3 weeks ago
0
add jsonlines
#192
gshennvm
closed
3 weeks ago
0
SteerLM 2.0
#191
yidong72
closed
2 weeks ago
0
add sft on dpo
#190
gshennvm
opened
3 weeks ago
0
change dev version
#189
gshennvm
closed
3 weeks ago
0
sync: main to dev
#188
github-actions[bot]
closed
3 weeks ago
0
Geshen/upgrade to 24 05
#187
gshennvm
closed
3 weeks ago
0
fix for issue 46
#186
guyknvda
opened
1 month ago
0
how to fine-tune models with multi-nodes
#185
panjianfei
closed
1 month ago
2
Ea/multiple validation sets
#184
eloialonso
opened
1 month ago
0
fix save interval in docs
#183
gshennvm
closed
1 month ago
0
SPIN TRT Integration
#182
trias702
closed
2 weeks ago
0
add packed dataset
#181
gshennvm
opened
1 month ago
1
draft for steerlm 2.0
#180
yidong72
closed
3 weeks ago
0
sync: main to dev
#179
github-actions[bot]
closed
1 month ago
0
delete sampler since it's merged into nemo
#178
gshennvm
closed
1 month ago
0
Fix max sample length in RLHF dataset
#177
odelalleau
opened
1 month ago
0
Some code related to `train_valid_test_num_samples` may be wrong / unused
#176
odelalleau
opened
1 month ago
0
how to fine-tune Qwen1.5 models based on Nemo
#175
panjianfei
closed
1 month ago
4
Next