Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion trl/trainer/sft_config.py
Original file line number Diff line number Diff line change
Expand Up @@ -49,7 +49,8 @@ class SFTConfig(TrainingArguments):
The number of sequences to use for the `ConstantLengthDataset`. Defaults to `1024`.
chars_per_token (`Optional[float]`):
The number of characters per token to use for the `ConstantLengthDataset`. Defaults to `3.6`. You can check how this is computed in the
stack-llama example: https://github.com/huggingface/trl/blob/08f550674c553c36c51d1027613c29f14f3676a5/examples/stack_llama/scripts/supervised_finetuning.py#L53.
stack-llama example:
[chars_token_ratio](https://github.com/huggingface/trl/blob/08f550674c553c36c51d1027613c29f14f3676a5/examples/stack_llama/scripts/supervised_finetuning.py#L53).
"""

dataset_text_field: Optional[str] = None
Expand Down