prompt_strategies.dpo.chat_template
prompt_strategies.dpo.chat_template
DPO prompt strategies for using tokenizer chat templates.
Functions
| Name | Description |
|---|---|
| argilla_chat | DPO chat template strategy for argilla-style datasets. |
argilla_chat
prompt_strategies.dpo.chat_template.argilla_chat(cfg, dataset_idx=0, **kwargs)DPO chat template strategy for argilla-style datasets.
For argilla-style datasets where chosen/rejected contain full conversations instead of single response messages. Extracts the conversation history from the chosen field and formats both chosen/rejected responses using the configured chat template.
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
| cfg | Configuration object containing chat_template and dataset settings | required | |
| dataset_idx | Index of the dataset in the config (default: 0) | 0 |
|
| **kwargs | Additional keyword arguments (unused) | {} |
Returns
| Name | Type | Description |
|---|---|---|
| tuple | (transform_fn, dataset_kwargs) where: - transform_fn: Function to transform dataset samples - dataset_kwargs: Dict with ‘remove_columns’ specifying columns to drop |
Dataset format
{ “chosen”: [ {“role”: “user”, “content”: “…”}, {“role”: “assistant”, “content”: “…”} ], “rejected”: [ {“role”: “user”, “content”: “…”}, {“role”: “assistant”, “content”: “…”} ] }