Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
OLMo-core v2.5.0
Light Logo Dark Logo
OLMo-core v2.5.0

Overview

  • Introduction
  • Installation

Guides

  • All-in-one for researchers
  • Datasets and Data loading
  • Data Mixing
  • Native Generation and Chat

Examples

  • HuggingFace models
  • Train an LLM

API Reference

  • config
  • data
    • data.numpy_dataset
    • data.source_mixture
    • data.collator
    • data.composable
    • data.mixes
    • data.tokenizer
    • data.data_loader
    • data.types
    • data.utils
  • distributed
    • distributed.checkpoint
    • distributed.parallel
    • distributed.utils
  • eval
    • eval.metrics
    • eval.evaluator
    • eval.lm_evaluator
  • exceptions
  • float8
  • fs_cache
  • generate
    • generate.generation_module
    • generate.sampling
    • generate.utils
  • io
  • launch
  • model_ladder
  • nn
    • nn.attention
    • nn.conversion
    • nn.feed_forward
    • nn.functional
    • nn.hf
    • nn.layer_norm
    • nn.lm_head
    • nn.moe
    • nn.rope
    • nn.transformer
  • optim
  • testing
  • train
    • train.callbacks
    • train.train_module
  • utils

Development

  • License
  • CHANGELOG
  • GitHub Repository
Back to top
View this page

generateΒΆ

Submodules

  • generate.generation_module
    • GenerationConfig
    • GenerationModule
    • TransformerGenerationModule
    • TransformerGenerationModuleConfig
  • generate.sampling
    • greedy_selection()
    • top_k_filtering()
    • top_p_filtering()
    • select_next_token()
  • generate.utils
    • selective_log_softmax()
Next
generate.generation_module
Previous
fs_cache
Copyright © 2026, Allen Institute for Artificial Intelligence
Made with Sphinx and @pradyunsg's Furo