Save consolidated weights and copy configuration file Args: consolidated_weights: Dictionary of consolidated parameters conf

Set the state dict of the Peft model. Args: model ([`PeftModel`]): The Peft model. peft_model_state_dict (`dict`): The state

Shift input ids one token to the right. Args: input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): input ids

This stores parameter gradients of past tasks. pp: parameters grads: gradients grad_dims: list with n

Update similarity tensor for selected task indices

Forward missing attributes to the wrapped module.

Forward missing attributes to the wrapped module.

Forward missing attributes to the wrapped module.

Forward missing attributes to the wrapped module.

:param depth: current depth :param grads_tensor: (num_tasks, lora_depth, feature_dim)

Initialize the KD-tree. Args: points (Tensor): A tensor of shape (N, D) where N is the number of points and D is the dim

Yields examples.

Returns SplitGenerators.