maxent_grpo.training.rollout.context

Shared generation context dataclass used by local and vLLM paths.

Classes

GenerationContext(max_prompt_len, ...[, ...])

Configuration required to produce completions for each training batch.

class maxent_grpo.training.rollout.context.GenerationContext(max_prompt_len, max_completion_len, gen_temperature, gen_top_p, use_vllm, vllm, accelerator, model, tokenizer, generation_stats, device, penalty=<factory>, prompt_char_limit=None, *, vllm_mode='server')

Bases: GenerationPenaltyPassthroughMixin, GenerationSamplingConfig

Configuration required to produce completions for each training batch.

Parameters:

• max_prompt_len (int)

• max_completion_len (int)

• gen_temperature (float)

• gen_top_p (float)

• use_vllm (bool)

• vllm (VLLMClientConfig)

• accelerator (Accelerator)

• model (PreTrainedModel)

• tokenizer (PreTrainedTokenizer)

• generation_stats (Dict[str, int])

• device (Any)

• penalty (GenerationPenaltyConfig)

• prompt_char_limit (int | None)

• vllm_mode (str)

accelerator: TypesAccelerator

model: TypesPreTrainedModel

tokenizer: TypesPreTrainedTokenizer

generation_stats: Dict[str, int]

device: Any

penalty: GenerationPenaltyConfig

prompt_char_limit: int | None = None

as_dict()

Return a lightweight representation useful for logging/debugging.

Return type:

Dict[str, Any]

max_prompt_len: int

max_completion_len: int

gen_temperature: float

gen_top_p: float

use_vllm: bool

vllm: VLLMClientConfig