issues
search
NVIDIA
/
NeMo-Aligner
Scalable toolkit for efficient model alignment
Apache License 2.0
625
stars
78
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
docs: fix code block rendering
#369
ashors1
closed
3 weeks ago
0
mamba sft support
#368
haifengqian
closed
1 week ago
0
docs: main readme and sft docs
#367
okuchaiev
closed
3 weeks ago
0
ci: Sign-off cherry pick
#366
ko3n1g
closed
3 weeks ago
0
Cherry pick `docs: 0.5.0 documentation updates (346)` into `r0.5.0`
#365
ko3n1g
closed
3 weeks ago
0
Make dev dockerfile compatible with dev branch
#364
ashors1
closed
3 weeks ago
1
sync: main to dev
#363
github-actions[bot]
closed
3 weeks ago
0
fix: resolve distributed test hangs
#362
ashors1
closed
3 weeks ago
0
feat: update reward model to support scaled and margin BT
#361
Zhilin123
closed
3 weeks ago
7
How can I use nvidia/Llama-3.1-Nemotron-70B-Reward-HF directly for inference?
#360
arunasank
opened
1 month ago
4
ci(fix): Make pre-commit sign so that a rebase isn't required
#359
terrykong
closed
1 month ago
0
build: use cached trtllm build when updating aligner tag
#358
gwarmstrong
closed
3 weeks ago
0
feat: adds REINFORCE algorithm
#357
abukharin3
closed
1 day ago
0
ci: re-enables unit tests by marking broken tests and require all tests
#356
terrykong
closed
1 month ago
0
ci: Allow building with merge-commit on forks
#355
ko3n1g
closed
1 month ago
0
ci: support building on forks
#354
terrykong
closed
1 month ago
1
ci: Validate PR title
#353
ko3n1g
closed
1 month ago
0
fix: correct batch tokenization when sequence exceeds encoder length
#352
gwarmstrong
closed
1 month ago
1
serve_reward_model goes down
#351
AtsunoriFujita
opened
1 month ago
3
Fixed attribute_annotate.py is not worked by KeyError: 'exceeded' #349
#350
AtsunoriFujita
opened
1 month ago
0
`attribute_annotate.py` is not worked by KeyError: 'exceeded'
#349
AtsunoriFujita
opened
1 month ago
0
feat: add knowledge distillation support for SFT
#348
ashors1
closed
2 weeks ago
0
Cherry pick `docs: Added note about GBS and jsonl samples to DPO tutorial (345)` into `r0.5.0`
#347
ko3n1g
closed
1 month ago
0
docs: 0.5.0 documentation updates
#346
ashors1
closed
3 weeks ago
0
docs: Added note about GBS and jsonl samples to DPO tutorial
#345
trias702
closed
1 month ago
0
Add DPO and PPO presubmit tests
#344
ashors1
closed
1 month ago
0
ci: Update runner labels
#343
ko3n1g
closed
1 month ago
0
Unable to pip install nemo-aligner
#342
SCccc21
opened
1 month ago
1
Dev/linky data
#341
yangchao-zhou
opened
1 month ago
0
[Question] Converting a Megatron-LM ckpt to nemo so we can use NeMo-Aligner for post-training
#340
abgoswam
opened
1 month ago
0
feat: support mcore optimizers
#339
terrykong
closed
1 month ago
0
Fixed string literal error
#338
SulRash
opened
1 month ago
1
sync: main to dev
#337
github-actions[bot]
closed
4 weeks ago
0
LD_LIBRARY_PATH override in dockerfile causes failure in CI
#336
terrykong
closed
1 month ago
4
Cherry pick `fix: serve_reward_model.py no longer errors if pretrained GBS indivisible by parallel state (333)` into `r0.5.0`
#335
ko3n1g
closed
1 month ago
0
sync: main to dev
#334
github-actions[bot]
closed
1 month ago
0
fix: serve_reward_model.py no longer errors if pretrained GBS indivisible by parallel state
#333
terrykong
closed
1 month ago
0
docs: adds a known_errors.rst to improve UX
#332
terrykong
opened
1 month ago
1
build: Build trtllm with multi-stage
#331
ko3n1g
closed
2 weeks ago
1
fix: unit test imports
#330
terrykong
closed
1 month ago
2
Cherry pick feat: Adds Rejection Sampling Algorithm
#329
terrykong
closed
1 month ago
0
ci: Use Nemo FW templates
#328
ko3n1g
closed
2 weeks ago
0
sync: main to dev
#327
github-actions[bot]
closed
1 month ago
0
Cherry pick `docs: Add sidebar headings for RPO/IPO/DPO with LoRA (319)` into `r0.5.0`
#326
ko3n1g
closed
1 month ago
0
Cherry pick `docs: Fixes performance table for rlhf (322)` into `r0.5.0`
#325
ko3n1g
closed
1 month ago
0
Cherry pick `fix: Dockerfile now correctly updates aligner in incremental aligner-bump target (323)` into `r0.5.0`
#324
ko3n1g
closed
1 month ago
0
fix: Dockerfile now correctly updates aligner in incremental aligner-bump target
#323
terrykong
closed
1 month ago
0
docs: Fixes performance table for rlhf
#322
terrykong
closed
1 month ago
0
feat: Self-Rewarding Algorithm with TRT Support
#321
trias702
opened
1 month ago
0
feat: Upgrading TRTLLM to v13
#320
terrykong
closed
3 weeks ago
1
Previous
Next