r""" Instantiates a new model from a pretrained model from `transformers`. The pretrained model is loaded using the `from_pretrained`

r""" Push the pretrained model to the hub. This method is a wrapper around `transformers.PreTrainedModel.push_to_hub`. Please refer to

Updates running moments from batch's moments computed across ranks

Generates model completions for the provided prompts. Args: prompts (`list[str]`): List of text prompts

Log `logs` on the various objects watching training, including stored metrics. Args: logs (`dict[str, float]`):

Pads a list of tensors to the same shape along the first dimension. Args: tensors (`list[torch.Tensor]`): List of input

Empties the cache of the available torch device. This function checks for the availability of different torch devices (XPU, MLU, NPU, CUDA) and e

Generate a `ModelCard` from a template. Args: base_model (`str` or `None`): Base model name. model_name (`str`):

Parse command-line args and config file into instances of the specified dataclass types. This method wraps [`transformers.HfArgument

If Comet integration is enabled, return the URL of the current Comet experiment; otherwise, return `None`.

r""" Check if the example is in a conversational format. Args: example (`dict[str, Any]`): A single data entry of a datas

A memory-efficient implementation of the common `log_softmax -> gather` operation. This function is equivalent to the following naive implem

Context manager to unwrap distributed or accelerated models for generation tasks. Args: model (`Union[DistributedDataParallel, DeepS

Prepares the model for training with DeepSpeed (both for stage 2 and 3), configuring the appropriate settings based on the model and batch si

If Comet integration is enabled logs a table to the Comet experiment if it is currently running. Args: name (`str`): Tab

r""" Save the pretrained model to a directory. This method is a wrapper around `transformers.PreTrainedModel.save_pretrained`. Please

Returns the tokenizer used for tokenizing text inputs

Computes the reward logits and the rewards for a given model and query responses. Args: model (`torch.nn.Module`): The m

Perform a single step of training. Args: epoch (int): The current epoch. global_step (int): The current glob

Compute mean of tensor with a masked values.

Creates a static reference copy of a model. Note that model will be in `.eval()` mode. Args: model (`PreTrainedModelWrapper`): The m

Shift non-zero elements in the mask and corresponding tensors to the left. This function operates on a binary mask and any number of additio

r""" Return the state_dict of the pretrained model.

Takes an N-dimensional bool tensor and returns an (N-1)-dimensional tensor of integers giving the position of the first True in each "row".

Performs a forward pass through the model with the given query responses and pad token ID. Args: model (`torch.nn.Module`):

Context manager for handling null reference model (that is, peft adapter manipulation).

A context manager function for profiling a block of code. Results are logged to Weights & Biases or MLflow depending on the trainer's configu

Truncates the input tensor from the right side after the first occurrence of the stop token. Args: input_ids (`torch.Tensor`):

Updates a specific named parameter in the model and broadcasts it to other processes. Args: name (`str`):

Compute log-probs and (optionally) entropies for each token.

Compute the log probabilities of the given labels under the given logits. Args: logits: Logits of the model (unno

r""" If the example is in a conversational format, apply a chat template to it. Args: example (`dict[str, list[dict[str, str]]`):

r""" Post initialization method. This method is called after the model is instantiated and loaded from a checkpoint. It can be used to

Setup chat format by adding special tokens to the tokenizer, setting the correct format, and extending the embedding layer of the model based

Returns the text encoder used for encoding text inputs

Truncates the responses at the first occurrence of the stop token, filling the rest with pad tokens. Args: stop_token_id (`int`):

Returns the 2d U-Net model used for diffusion.

Compute KL log probabilities for a given batch.

Generates completions for a list of pre-formatted prompts from the given model. Args: prompts (list[str]): A list of input prompts f

r""" Get the current device. For GPU & XPU, we return the local process index using the `accelerate.PartialState` object to handle cor

As opposed to the default direction of broadcasting (right to left), this function broadcasts from left to right Args: input

Separate the kwargs from the arguments that we support inside `supported_args` and the ones that we don't.

Append a new segment to the history. Args: text (`str`): The text of the new segment. tokens (`torch.LongTen

Returns the autocast context manager

Computes log probabilities of the reference model for a single padded batch of a DPO specific dataset.

Runs the given model on the given batch of inputs, concatenating the chosen and rejected inputs together. We do this to avoid doing

Return a list of indices so that each slice of `batch_size` consecutive indices correspond to elements of similar lengths. To do this, the in