KeyError in DPO Trainer, evaluation_loop

[qingjianbuyi]

System Info

• TRL version: 0.12.1

Information

• The official example scripts
• My own modified scripts

Tasks

• An officially supported task in the examples folder
• My own task or dataset (give details below)

Reproduction

python examples/scripts/dpo.py \
    --dataset_name trl-lib/ultrafeedback_binarized \
    --model_name_or_path Qwen/Qwen2-0.5B-Instruct \
    --learning_rate 5.0e-6 \
    --num_train_epochs 1 \
    --per_device_train_batch_size 2 \
    --gradient_accumulation_steps 8 \
    --gradient_checkpointing \
    --logging_steps 25 \
    --eval_strategy steps \
    --eval_steps 50 \
    --output_dir Qwen2-0.5B-DPO \
    --no_remove_unused_columns \
    --use_peft \
    --lora_r 32 \
    --lora_alpha 16 \
    --generate_during_eval

Expected behavior

            self.log(
                {
                    "game_log": wandb.Table(
                        columns=["Prompt", "Policy", "Ref Model"],
                        rows=[
                            [prompt, pol[len(prompt) :], ref[len(prompt) :]]
                            for prompt, pol, ref in zip(
#                                random_batch["prompt"], policy_output_decoded, ref_output_decoded
                                self.tokenizer.decode(random_batch["prompt_input_ids"]), policy_output_decoded, ref_output_decoded
                            )
                        ],
                    )
                }
            )

Checklist

• I have checked that my issue isn't already filed (see open issues)
• I have included my system information
• Any code provided is minimal, complete, and reproducible (more on MREs)
• Any code provided is properly formatted in code blocks, (no screenshot, more on code blocks)
• Any traceback provided is complete