CPO Trainer

Overview

Contrastive Preference Optimization (CPO) as introduced in the paper Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation by Haoran Xu, Amr Sharaf, Yunmo Chen, Weiting Tan, Lingfeng Shen, Benjamin Van Durme, Kenton Murray, and Young Jin Kim. At a high-level, CPO trains models to avoid generating adequate, but not perfect translations in Machine Translation (MT) tasks. However, CPO is a general approximation to the DPO loss and can be applied to other domains like chat.

CPO aims to mitigate two fundamental shortcomings of SFT. First, SFT’s methodology of minimizing the discrepancy between predicted outputs and gold-standard references inherently caps model performance at the quality level of the training data. Secondly, SFT lacks a mechanism to prevent the model from rejecting mistakes in translations. The CPO objective is derived from the DPO objective.

Quick start

This example demonstrates how to train a model using the CPO method. We use the Qwen 0.5B model as the base model. We use the preference data from the UltraFeedback dataset. You can view the data in the dataset here:

Below is the script to train the model:

# train_cpo.py
from datasets import load_dataset
from trl import CPOConfig, CPOTrainer
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2-0.5B-Instruct")
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2-0.5B-Instruct")
train_dataset = load_dataset("trl-lib/ultrafeedback_binarized", split="train")

training_args = CPOConfig(output_dir="Qwen2-0.5B-CPO", logging_steps=10)
trainer = CPOTrainer(model=model, args=training_args, processing_class=tokenizer, train_dataset=train_dataset)
trainer.train()

Execute the script using the following command:

accelerate launch train_cpo.py

Expected dataset type

CPO requires a preference dataset. The CPOTrainer supports both conversational and standard dataset format. When provided with a conversational dataset, the trainer will automatically apply the chat template to the dataset.

Example script

We provide an example script to train a model using the CPO method. The script is available in examples/scripts/cpo.py

To test the CPO script with the Qwen2 0.5B model on the UltraFeedback dataset, run the following command:

accelerate launch examples/scripts/cpo.py \
    --model_name_or_path Qwen/Qwen2-0.5B-Instruct \
    --dataset_name trl-lib/ultrafeedback_binarized \
    --num_train_epochs 1 \
    --logging_steps 25 \
    --output_dir Qwen2-0.5B-CPO

Logged metrics

While training and evaluating we record the following reward metrics:

rewards/chosen: the mean log probabilities of the policy model for the chosen responses scaled by beta
rewards/rejected: the mean log probabilities of the policy model for the rejected responses scaled by beta
rewards/accuracies: mean of how often the chosen rewards are > than the corresponding rejected rewards
rewards/margins: the mean difference between the chosen and corresponding rejected rewards
nll_loss: the mean negative log likelihood loss of the policy model for the chosen responses

CPO variants

Simple Preference Optimization (SimPO)

The SimPO method is also implemented in the CPOTrainer. SimPO is an alternative loss that adds a reward margin, allows for length normalization, and does not use BC regularization. To use this loss, we can use SimPO easily by turning on loss_type="simpo" and cpo_alpha=0 in the CPOConfig.

CPO-SimPO

We also offer the combined use of CPO and SimPO, which enables more stable training and improved performance. Learn more details at CPO-SimPO GitHub. To use this method, simply enable SimPO by setting loss_type="simpo" and a non-zero cpo_alpha in the CPOConfig.

Loss functions

The CPO algorithm supports several loss functions. The loss function can be set using the loss_type parameter in the CPOConfig. The following loss functions are supported:

`loss_type=`	Description
`"sigmoid"` (default)	Given the preference data, we can fit a binary classifier according to the Bradley-Terry model and in fact the DPO authors propose the sigmoid loss on the normalized likelihood via the `logsigmoid` to fit a logistic regression.
`"hinge"`	The RSO authors propose to use a hinge loss on the normalized likelihood from the SLiC paper. In this case, the `beta` is the reciprocal of the margin.
`"ipo"`	The IPO authors provide a deeper theoretical understanding of the DPO algorithms and identify an issue with overfitting and propose an alternative loss. In this case, the `beta` is the reciprocal of the gap between the log-likelihood ratios of the chosen vs the rejected completion pair and thus the smaller the `beta` the larger this gaps is. As per the paper the loss is averaged over log-likelihoods of the completion (unlike DPO which is summed only).

For Mixture of Experts Models: Enabling the auxiliary loss

MOEs are the most efficient if the load is about equally distributed between experts.
To ensure that we train MOEs similarly during preference-tuning, it is beneficial to add the auxiliary loss from the load balancer to the final loss.

This option is enabled by setting output_router_logits=True in the model config (e.g. MixtralConfig).
To scale how much the auxiliary loss contributes to the total loss, use the hyperparameter router_aux_loss_coef=... (default: 0.001) in the model config.

CPOTrainer

class trl.CPOTrainer

< source >

( model: Union = None args: Optional = None data_collator: Optional = None train_dataset: Optional = None eval_dataset: Union = None processing_class: Union = None model_init: Optional = None callbacks: Optional = None optimizers: Tuple = (None, None) preprocess_logits_for_metrics: Optional = None peft_config: Optional = None compute_metrics: Optional = None )

Parameters

model (transformers.PreTrainedModel) — The model to train, preferably an AutoModelForSequenceClassification.
args (CPOConfig) — The CPO config arguments to use for training.
data_collator (transformers.DataCollator) — The data collator to use for training. If None is specified, the default data collator (DPODataCollatorWithPadding) will be used which will pad the sequences to the maximum length of the sequences in the batch, given a dataset of paired sequences.
train_dataset (datasets.Dataset) — The dataset to use for training.
eval_dataset (datasets.Dataset) — The dataset to use for evaluation.
processing_class (PreTrainedTokenizerBase or BaseImageProcessor or FeatureExtractionMixin or ProcessorMixin, optional) — Processing class used to process the data. If provided, will be used to automatically process the inputs for the model, and it will be saved along the model to make it easier to rerun an interrupted training or reuse the fine-tuned model.
model_init (Callable[[], transformers.PreTrainedModel]) — The model initializer to use for training. If None is specified, the default model initializer will be used.
callbacks (List[transformers.TrainerCallback]) — The callbacks to use for training.
optimizers (Tuple[torch.optim.Optimizer, torch.optim.lr_scheduler.LambdaLR]) — The optimizer and scheduler to use for training.
preprocess_logits_for_metrics (Callable[[torch.Tensor, torch.Tensor], torch.Tensor]) — The function to use to preprocess the logits before computing the metrics.
peft_config (Dict, defaults to None) — The PEFT configuration to use for training. If you pass a PEFT configuration, the model will be wrapped in a PEFT model.
compute_metrics (Callable[[EvalPrediction], Dict], optional) — The function to use to compute the metrics. Must take a EvalPrediction and return a dictionary string to metric values.

Initialize CPOTrainer.

build_tokenized_answer

< source >

( prompt answer )

Llama tokenizer does satisfy enc(a + b) = enc(a) + enc(b). It does ensure enc(a + b) = enc(a) + enc(a + b)[len(enc(a)):]. Reference: https://github.com/EleutherAI/lm-evaluation-harness/pull/531#issuecomment-1595586257

concatenated_forward

< source >

( model: Module batch: Dict )

Run the given model on the given batch of inputs, concatenating the chosen and rejected inputs together.

We do this to avoid doing two forward passes, because it’s faster for FSDP.

concatenated_inputs

< source >

( batch: Dict is_encoder_decoder: bool = False label_pad_token_id: int = -100 padding_value: int = 0 device: Optional = None )

Concatenate the chosen and rejected inputs into a single tensor.

cpo_loss

< source >

( policy_chosen_logps: FloatTensor policy_rejected_logps: FloatTensor ) → A tuple of three tensors

Returns

A tuple of three tensors

(losses, chosen_rewards, rejected_rewards). The losses tensor contains the CPO loss for each example in the batch. The chosen_rewards and rejected_rewards tensors contain the rewards for the chosen and rejected responses, respectively.

Compute the CPO loss for a batch of policy and reference model log probabilities.

create_model_card

< source >

( model_name: Optional = None dataset_name: Optional = None tags: Union = None )

Parameters

model_name (str, optional, defaults to None) — The name of the model.
dataset_name (str, optional, defaults to None) — The name of the dataset used for training.
tags (str, List[str] or None, optional, defaults to None) — Tags to be associated with the model card.

Creates a draft of a model card using the information available to the Trainer.

evaluation_loop

< source >

( dataloader: DataLoader description: str prediction_loss_only: Optional = None ignore_keys: Optional = None metric_key_prefix: str = 'eval' )

Overriding built-in evaluation loop to store metrics for each batch. Prediction/evaluation loop, shared by Trainer.evaluate() and Trainer.predict().

Works both with or without labels.

generate_from_model

< source >

( model batch: Dict )

Generate samples from the model and reference model for the given batch of inputs.

get_batch_logps

< source >

( logits: FloatTensor labels: LongTensor average_log_prob: bool = False label_pad_token_id: int = -100 is_encoder_decoder: bool = False )

Compute the log probabilities of the given labels under the given logits.

get_batch_loss_metrics

< source >

( model batch: Dict train_eval: Literal = 'train' )

Compute the CPO loss and other metrics for the given batch of inputs for train or test.

log

< source >

( logs: Dict )

Parameters

logs (Dict[str, float]) — The values to log.

Log logs on the various objects watching training, including stored metrics.

tokenize_row

< source >

( feature model: Union = None )

Tokenize a single row from a CPO specific dataset.

At this stage, we don’t convert to PyTorch tensors yet; we just handle the truncation in case the prompt + chosen or prompt + rejected responses is/are too long. First we truncate the prompt; if we’re still too long, we truncate the chosen/rejected.

We also create the labels for the chosen/rejected responses, which are of length equal to the sum of the length of the prompt and the chosen/rejected response, with label_pad_token_id for the prompt tokens.

CPOConfig

class trl.CPOConfig

< source >

( output_dir: str overwrite_output_dir: bool = False do_train: bool = False do_eval: bool = False do_predict: bool = False eval_strategy: Union = 'no' prediction_loss_only: bool = False per_device_train_batch_size: int = 8 per_device_eval_batch_size: int = 8 per_gpu_train_batch_size: Optional = None per_gpu_eval_batch_size: Optional = None gradient_accumulation_steps: int = 1 eval_accumulation_steps: Optional = None eval_delay: Optional = 0 torch_empty_cache_steps: Optional = None learning_rate: float = 1e-06 weight_decay: float = 0.0 adam_beta1: float = 0.9 adam_beta2: float = 0.999 adam_epsilon: float = 1e-08 max_grad_norm: float = 1.0 num_train_epochs: float = 3.0 max_steps: int = -1 lr_scheduler_type: Union = 'linear' lr_scheduler_kwargs: Union = <factory> warmup_ratio: float = 0.0 warmup_steps: int = 0 log_level: Optional = 'passive' log_level_replica: Optional = 'warning' log_on_each_node: bool = True logging_dir: Optional = None logging_strategy: Union = 'steps' logging_first_step: bool = False logging_steps: float = 500 logging_nan_inf_filter: bool = True save_strategy: Union = 'steps' save_steps: float = 500 save_total_limit: Optional = None save_safetensors: Optional = True save_on_each_node: bool = False save_only_model: bool = False restore_callback_states_from_checkpoint: bool = False no_cuda: bool = False use_cpu: bool = False use_mps_device: bool = False seed: int = 42 data_seed: Optional = None jit_mode_eval: bool = False use_ipex: bool = False bf16: bool = False fp16: bool = False fp16_opt_level: str = 'O1' half_precision_backend: str = 'auto' bf16_full_eval: bool = False fp16_full_eval: bool = False tf32: Optional = None local_rank: int = -1 ddp_backend: Optional = None tpu_num_cores: Optional = None tpu_metrics_debug: bool = False debug: Union = '' dataloader_drop_last: bool = False eval_steps: Optional = None dataloader_num_workers: int = 0 dataloader_prefetch_factor: Optional = None past_index: int = -1 run_name: Optional = None disable_tqdm: Optional = None remove_unused_columns: Optional = True label_names: Optional = None load_best_model_at_end: Optional = False metric_for_best_model: Optional = None greater_is_better: Optional = None ignore_data_skip: bool = False fsdp: Union = '' fsdp_min_num_params: int = 0 fsdp_config: Union = None fsdp_transformer_layer_cls_to_wrap: Optional = None accelerator_config: Union = None deepspeed: Union = None label_smoothing_factor: float = 0.0 optim: Union = 'adamw_torch' optim_args: Optional = None adafactor: bool = False group_by_length: bool = False length_column_name: Optional = 'length' report_to: Union = None ddp_find_unused_parameters: Optional = None ddp_bucket_cap_mb: Optional = None ddp_broadcast_buffers: Optional = None dataloader_pin_memory: bool = True dataloader_persistent_workers: bool = False skip_memory_metrics: bool = True use_legacy_prediction_loop: bool = False push_to_hub: bool = False resume_from_checkpoint: Optional = None hub_model_id: Optional = None hub_strategy: Union = 'every_save' hub_token: Optional = None hub_private_repo: bool = False hub_always_push: bool = False gradient_checkpointing: bool = False gradient_checkpointing_kwargs: Union = None include_inputs_for_metrics: bool = False include_for_metrics: List = <factory> eval_do_concat_batches: bool = True fp16_backend: str = 'auto' evaluation_strategy: Union = None push_to_hub_model_id: Optional = None push_to_hub_organization: Optional = None push_to_hub_token: Optional = None mp_parameters: str = '' auto_find_batch_size: bool = False full_determinism: bool = False torchdynamo: Optional = None ray_scope: Optional = 'last' ddp_timeout: Optional = 1800 torch_compile: bool = False torch_compile_backend: Optional = None torch_compile_mode: Optional = None dispatch_batches: Optional = None split_batches: Optional = None include_tokens_per_second: Optional = False include_num_input_tokens_seen: Optional = False neftune_noise_alpha: Optional = None optim_target_modules: Union = None batch_eval_metrics: bool = False eval_on_start: bool = False use_liger_kernel: Optional = False eval_use_gather_object: Optional = False average_tokens_across_devices: Optional = False max_length: Optional = None max_prompt_length: Optional = None max_completion_length: Optional = None beta: float = 0.1 label_smoothing: float = 0.0 loss_type: Literal = 'sigmoid' disable_dropout: bool = True cpo_alpha: float = 1.0 simpo_gamma: float = 0.5 label_pad_token_id: int = -100 padding_value: Optional = None truncation_mode: str = 'keep_end' generate_during_eval: bool = False is_encoder_decoder: Optional = None model_init_kwargs: Optional = None dataset_num_proc: Optional = None )

Parameters

learning_rate (float, optional, defaults to 1e-6) — Initial learning rate for AdamW optimizer. The default value replaces that of TrainingArguments.
max_length (Optional[int], optional, defaults to None) — Maximum length of the sequences (prompt + completion) in the batch. This argument is required if you want to use the default data collator.
max_prompt_length (Optional[int], optional, defaults to None) — Maximum length of the prompt. This argument is required if you want to use the default data collator.
max_completion_length (Optional[int], optional, defaults to None) — Maximum length of the completion. This argument is required if you want to use the default data collator and your model is an encoder-decoder.
beta (float, optional, defaults to 0.1) — Parameter controlling the deviation from the reference model. Higher β means less deviation from the reference model. For the IPO loss (loss_type="ipo"), β is the regularization parameter denoted by τ in the paper.
label_smoothing (float, optional, defaults to 0.0) — Label smoothing factor. This argument is required if you want to use the default data collator.
loss_type (str, optional, defaults to "sigmoid") — Type of loss to use. Possible values are:
- "sigmoid": sigmoid loss from the original DPO paper.
- "hinge": hinge loss on the normalized likelihood from the SLiC paper.
- "ipo": IPO loss from the IPO paper.
- "simpo": SimPO loss from the SimPO paper.
disable_dropout (bool, optional, defaults to True) — Whether to disable dropout in the model.
cpo_alpha (float, optional, defaults to 1.0) — Weight of the BC regularizer in CPO training.
simpo_gamma (float, optional, defaults to 0.5) — Target reward margin for the SimPO loss, used only when the loss_type="simpo".
label_pad_token_id (int, optional, defaults to -100) — Label pad token id. This argument is required if you want to use the default data collator.
padding_value (Optional[int], optional, defaults to None) — Padding value to use. If None, the padding value of the tokenizer is used.
truncation_mode (str,optional, defaults to "keep_end") — Truncation mode to use when the prompt is too long. Possible values are "keep_end" or "keep_start". This argument is required if you want to use the default data collator.
generate_during_eval (bool, optional, defaults to False) — If True, generates and logs completions from the model to W&B during evaluation.
is_encoder_decoder (Optional[bool], optional, defaults to None) — When using the model_init argument (callable) to instantiate the model instead of the model argument, you need to specify if the model returned by the callable is an encoder-decoder model.
model_init_kwargs (Optional[Dict[str, Any]], optional, defaults to None) — Keyword arguments to pass to AutoModelForCausalLM.from_pretrained when instantiating the model from a string.
dataset_num_proc (Optional[int], optional, defaults to None) — Number of processes to use for processing the dataset.

Configuration class for the CPOTrainer.

Using HfArgumentParser we can turn this class into argparse arguments that can be specified on the command line.

< > Update on GitHub

TRL

CPO Trainer

Overview

Quick start

Expected dataset type

Example script

Logged metrics

CPO variants

Simple Preference Optimization (SimPO)

CPO-SimPO

Loss functions

For Mixture of Experts Models: Enabling the auxiliary loss

CPOTrainer

class trl.CPOTrainer

build_tokenized_answer

concatenated_forward

concatenated_inputs

cpo_loss

create_model_card

evaluation_loop

generate_from_model

get_batch_logps

get_batch_loss_metrics

log

tokenize_row

CPOConfig

class trl.CPOConfig