Triple

T18877030
Position Surface form Disambiguated ID Type / Status
Subject Tom Henighan E461713 entity
Predicate coAuthorOf P2389 FINISHED
Object Training Compute-Optimal Large Language Models NE NERFINISHED

How this triple was built (3 steps)

Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.

NER Named-entity recognition gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: Training Compute-Optimal Large Language Models | Statement: [Tom Henighan, coAuthorOf, Training Compute-Optimal Large Language Models]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Training Compute-Optimal Large Language Models
Context triple: [Tom Henighan, coAuthorOf, Training Compute-Optimal Large Language Models]
  • A. Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity
    "Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity" is a research paper that introduces a sparsely activated mixture-of-experts transformer architecture enabling efficient training and inference of language models with up to a trillion parameters.
  • B. Megatron-LM
    Megatron-LM is a large-scale language model training framework developed by NVIDIA, designed to efficiently train massive transformer models through model, tensor, and pipeline parallelism.
  • C. Language Models are Few-Shot Learners
    "Language Models are Few-Shot Learners" is a landmark research paper that demonstrated large-scale transformer-based language models can perform diverse tasks from just a few examples without task-specific training.
  • D. Language Models are Unsupervised Multitask Learners
    "Language Models are Unsupervised Multitask Learners" is a 2019 OpenAI research paper that demonstrated how large-scale unsupervised language models like GPT-2 can perform a wide range of tasks without task-specific training.
  • E. OPT: Open Pre-trained Transformer Language Models
    OPT: Open Pre-trained Transformer Language Models is a family of openly released large-scale transformer-based language models developed by Meta AI to provide transparent, reproducible alternatives to proprietary models like GPT-3.
  • F. None of above. chosen
  • G. Unsure - the case is ambiguous/there is not enough information to decide.
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07
Target entity: Training Compute-Optimal Large Language Models
Target entity description: "Training Compute-Optimal Large Language Models" is a research paper that analyzes how to most efficiently allocate computational resources when scaling large language models to maximize performance.
  • A. Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity
    "Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity" is a research paper that introduces a sparsely activated mixture-of-experts transformer architecture enabling efficient training and inference of language models with up to a trillion parameters.
  • B. Megatron-LM
    Megatron-LM is a large-scale language model training framework developed by NVIDIA, designed to efficiently train massive transformer models through model, tensor, and pipeline parallelism.
  • C. Language Models are Few-Shot Learners
    "Language Models are Few-Shot Learners" is a landmark research paper that demonstrated large-scale transformer-based language models can perform diverse tasks from just a few examples without task-specific training.
  • D. Language Models are Unsupervised Multitask Learners
    "Language Models are Unsupervised Multitask Learners" is a 2019 OpenAI research paper that demonstrated how large-scale unsupervised language models like GPT-2 can perform a wide range of tasks without task-specific training.
  • E. OPT: Open Pre-trained Transformer Language Models
    OPT: Open Pre-trained Transformer Language Models is a family of openly released large-scale transformer-based language models developed by Meta AI to provide transparent, reproducible alternatives to proprietary models like GPT-3.
  • F. None of above. chosen

Provenance (2 batches)

The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.

Step Stage Batch ID Status When
creating Elicitation batch_69d8dcfc3430819095ee6fc0eb4c06a5 completed April 10, 2026, 11:20 a.m.
NER Named-entity recognition batch_69e5c3cf0aa0819090991fc9e14910fb completed April 20, 2026, 6:12 a.m.
Created at: April 10, 2026, 11:57 a.m.