Skip to content

Actions: huggingface/trl

Actions

Hugging Face Issue Labeler

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
issue_auto_labeller.yml
725 workflow runs
725 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

why kl = nan when grpo train? Hugging Face Issue Labeler #676: Issue #4040 opened by uilstong
30s
30s
About "or None" and "defaults to None" Hugging Face Issue Labeler #675: Issue #4036 opened by qgallouedec
22s
22s
CI fails: ValueError: zero-size array to reduction operation maximum which has no identity Hugging Face Issue Labeler #674: Issue #4035 opened by albertvillanova
33s
33s
SFTTrainer with PEFT model Hugging Face Issue Labeler #673: Issue #4029 opened by lylaiyy
39s
39s
Abnormal results during DPO training Hugging Face Issue Labeler #672: Issue #4023 opened by wjjwyj
28s
28s
Does the reward function trainer support multimodal models now? Hugging Face Issue Labeler #671: Issue #4021 opened by hlp2020
21s
21s
No warning for unsupported int4 quantization Hugging Face Issue Labeler #670: Issue #4018 opened by MRiabov
39s
39s
ParallelismConfig not applied in GRPOTrainer: Trainer._prepare_context_parallel_inputs expects dict[torch.Tensor] but receives list[dict[list]] Hugging Face Issue Labeler #669: Issue #4016 opened by JdRion
22s
22s
Does the grpo vllm colocate training now support the InternVL3 8B model? Hugging Face Issue Labeler #668: Issue #4015 opened by DearFishi
23s
23s
current_gradient_accumulation_steps is undefined when eval_on_start==True Hugging Face Issue Labeler #667: Issue #4010 opened by konstantinjdobler
31s
31s
Can not from trl import DataCollatorForLanguageModeling Hugging Face Issue Labeler #666: Issue #4009 opened by lylaiyy
29s
29s
Training Step of GRPO in Wandb. Hugging Face Issue Labeler #665: Issue #4004 opened by mandyyyyii
25s
25s
DPO trainer with video content Hugging Face Issue Labeler #664: Issue #4002 opened by GabrieleGiudic
25s
25s
scale_rewards malfunctioned in GRPOTrainer Hugging Face Issue Labeler #663: Issue #3991 opened by Peter-Chou
25s
25s
accelerator.sync_gradients Hugging Face Issue Labeler #662: Issue #3988 opened by AriesJin
28s
28s
Feature Request: Save/Load Precomputed Ref Log-Probabilities in DPOTrainer Hugging Face Issue Labeler #661: Issue #3985 opened by ginkyenglee
31s
31s
GRPOTrainer uses undefined self.current_gradient_accumulation_steps, causing AttributeError during training Hugging Face Issue Labeler #660: Issue #3983 opened by ahatamiz
22s
22s
DPOTrainer uses tokenizer instead of processor for Gemma3 vision Hugging Face Issue Labeler #659: Issue #3982 opened by supreme-gg-gg
26s
26s
Extremely persistent error - ValueError: 'generation_batch_size' and 'steps_per_generation' can not be both configured at the same time Hugging Face Issue Labeler #658: Issue #3980 opened by Rakshith12-pixel
27s
27s
GRPOTrainer vLLM colocate hardcodes MASTER_PORT=12345 so no parallel runs possible Hugging Face Issue Labeler #657: Issue #3979 opened by ruggsea
31s
31s
kto trainer invalid configuration error Hugging Face Issue Labeler #656: Issue #3974 opened by bryanchrist
29s
29s
REQUEST: Dynamic Sampling for GRPO Hugging Face Issue Labeler #655: Issue #3973 opened by wenquanlu
23s
23s
REQUEST：Add estimation of flos metric in GRPO Trainer Hugging Face Issue Labeler #654: Issue #3967 opened by LLMOON
25s
25s
[Question] Why isn't vanilla REINFORCE implemented? Hugging Face Issue Labeler #653: Issue #3966 opened by nityadav
21s
21s
[QST] Colocation and resharding Hugging Face Issue Labeler #652: Issue #3963 opened by jeromeku
26s
26s