Search results

How can I specify a gpu id for vllm

⚡accelerateRelated to accelerateRelated to accelerate

🚀 deepspeedRelated to deepspeedRelated to deepspeed

Status: Open.

#3085 In huggingface/trl;

· YueChenkkk opened on Mar 14, 2025

Why the ratio in PPOTrainer use the same policy model rather than old and new policy model to generate?

⚡accelerateRelated to accelerateRelated to accelerate

🏋 PPORelated to PPORelated to PPO

Status: Open.

#3084 In huggingface/trl;

· Lynnzake opened on Mar 14, 2025

Why SFTTrainer process instruction data without EOS?

❓ questionSeeking clarification or more informationSeeking clarification or more information

🏋 SFTRelated to SFTRelated to SFT

Status: Open.

#3083 In huggingface/trl;

· DiaQusNet opened on Mar 14, 2025

Enabling GRPOConfig.include_tokens_per_second crashes GRPO training

⚡accelerateRelated to accelerateRelated to accelerate

🏋 GRPORelated to GRPORelated to GRPO

🐛 bugSomething isn't workingSomething isn't working

Status: Open.

#3081 In huggingface/trl;

· wizeng23 opened on Mar 14, 2025

DPO Trainer error "full() received an invalid combination of arguments"

🏋 DPORelated to DPORelated to DPO

🐛 bugSomething isn't workingSomething isn't working

Status: Open.

#3080 In huggingface/trl;

· sivaganesh07 opened on Mar 14, 2025

GRPO Trainer/Config Incomplete metrics logging to wandb

⚡accelerateRelated to accelerateRelated to accelerate

❓ questionSeeking clarification or more informationSeeking clarification or more information

🏋 GRPORelated to GRPORelated to GRPO

Status: Open.

#3078 In huggingface/trl;

· SpaceHunterInf opened on Mar 13, 2025

What does this line in the ppo trainer do ?

❓ questionSeeking clarification or more informationSeeking clarification or more information

🏋 PPORelated to PPORelated to PPO

Status: Open.

#3077 In huggingface/trl;

· LinHungShi opened on Mar 13, 2025

AutoModelForCausalLMWithValueHead initialize the model with deepspeed zero3

⚡ PEFTRelated to PEFTRelated to PEFT

⚡accelerateRelated to accelerateRelated to accelerate

🚀 deepspeedRelated to deepspeedRelated to deepspeed

Status: Open.

#3071 In huggingface/trl;

· JYX1216 opened on Mar 13, 2025

[Bug] ORPO Trainer hangs with multi-gpu on step 0

⚡accelerateRelated to accelerateRelated to accelerate

🏋 ORPORelated to ORPORelated to ORPO

🐛 bugSomething isn't workingSomething isn't working

Status: Open.

#3068 In huggingface/trl;

· NanoCode012 opened on Mar 13, 2025

🚀 Question: Is attention_mask still lower triangular when using packing=True in SFTTrainer?

⚡accelerateRelated to accelerateRelated to accelerate

❓ questionSeeking clarification or more informationSeeking clarification or more information

🏋 SFTRelated to SFTRelated to SFT

Status: Open.

#3067 In huggingface/trl;

· noforit opened on Mar 13, 2025

How to switch on the multi-GPU for GRPOTrainer?

🏋 GRPORelated to GRPORelated to GRPO

Status: Open.

#3066 In huggingface/trl;

· tjoymeed opened on Mar 13, 2025

Enable External Launcher Support for vLLM in TRL for Efficient GRPO Training

⚡accelerateRelated to accelerateRelated to accelerate

✨ enhancementNew feature or requestNew feature or request

🏋 GRPORelated to GRPORelated to GRPO

Status: Open.

#3064 In huggingface/trl;

· mtoslalibu opened on Mar 12, 2025