h] P S SK JrJr S SKJrJr S SKJr \ " S S\5 5 rg) ) dataclassfield)AnyOptional)TrainingArgumentsc . ^ \ rS rSr% Sr\R S/- r\" SSS0S9r\ \
S' \" S SS
0S9r \ \
S ' \" S SS
0S9r
\\ \
S' \" S SS0S9r\\\\4 \
S' \" S SS0S9r\\ \
S' \" SSS0S9r\\
S' \" SSS0S9r\ \
S' U 4S jrSrU =r$ )IterativeSFTConfig a
Configuration class for the [`IterativeSFTTrainer`].
The [`IterativeSFTTrainer`] is deprecated and will be removed in version 0.24.0. Please use the [`SFTTrainer`].
This class includes only the parameters that are specific to Iterative SFT training. For a full list of training
arguments, please refer to the [`~transformers.TrainingArguments`] documentation. Note that default values in this
class may differ from those in [`~transformers.TrainingArguments`].
Using [`~transformers.HfArgumentParser`] we can turn this class into
[argparse](https://docs.python.org/3/library/argparse#module-argparse) arguments that can be specified on the
command line.
Parameters:
> Parameters that control the model
model_init_kwargs (`dict[str, Any]` or `None`, *optional*, defaults to `None`):
Keyword arguments for [`~transformers.AutoModelForCausalLM.from_pretrained`], used when the `model`
argument of the [`IterativeSFTTrainer`] is provided as a string.
> Parameters that control the data preprocessing
max_length (`int` or `None`, *optional*, defaults to `None`):
Maximum length of the tokenized sequence. Sequences longer than `max_length` are truncated.
truncation_mode (`str`, *optional*, defaults to `"keep_end"`):
The truncation mode to use, either `"keep_end"` or `"keep_start"`.
optimize_device_cache (`bool`, *optional*, defaults to `False`):
Whether to optimize accelerator cache for slightly more memory-efficient training.
model_init_kwargs
helpzLog every X updates steps. Should be an integer or a float in range `[0,1)`. If smaller than 1, will be interpreted as ratio of total training steps.)defaultmetadata
logging_stepsTzZIf True, use gradient checkpointing to save memory at the expense of slower backward pass.gradient_checkpointingNzWhether to use bf16 (mixed) precision instead of 32-bit. Requires Ampere or higher NVIDIA architecture or Intel XPU or using CPU (use_cpu) or Ascend NPU. If not set, it defaults to `True` if `fp16` is not set.bf16zKeyword arguments for `AutoModelForCausalLM.from_pretrained`, used when the `model` argument of the `IterativeSFTTrainer` is provided as a string.z[Maximum length of the tokenized sequence. Sequences longer than `max_length` are truncated.
max_lengthkeep_endz>The truncation mode to use, either 'keep_end' or 'keep_start'.truncation_modeFzRWhether to optimize accelerator cache for slightly more memory-efficient training.optimize_device_cachec > U R c U R ( + O U R U l [ TU ]
5 U R S; a [ SU R 35 eg )N)r
keep_startz?truncation_mode must be either 'keep_end' or 'keep_start', got )r fp16super
__post_init__r
ValueError)self __class__s Z/home/james-whalen/.local/lib/python3.13/site-packages/trl/trainer/iterative_sft_config.pyr IterativeSFTConfig.__post_init__k sZ '+yy'8Odii
'A A ^_c_s_s^t uv v B )r )__name__
__module__ __qualname____firstlineno____doc__r _VALID_DICT_FIELDSr r float__annotations__r boolr r r dictstrr r intr r r __static_attributes__
__classcell__)r s @r r r s/ D +==AT@UU ! D
M5 $) p
$D ! !
D(4. 38 A
3xS#X/ !& q
!J
!Z[ OS #(no#4
w wr! r N) dataclassesr r typingr r transformersr r r! r r4 s2 ) * [w* [w [wr!