© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 95854
    Prev
    Next
    INDEX
    Explanations

    people's previous roles or accomplishments

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires strongly on person names (proper nouns referring to individuals).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    proper nouns referring to people, particularly names and titles of individuals mentioned in news articles.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    descriptive appositions and relative clauses that elaborate on a named entity, typically set off by commas and providing biographical or defining details.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     blijft
    -1.03
     chanteuse
    -1.02
     krijgt
    -0.96
    derabad
    -0.93
    Presenta
    -0.91
    σης
    -0.90
    终于
    -0.90
    felves
    -0.89
    Amicalement
    -0.89
     Áng
    -0.88
    POSITIVE LOGITS
     previously
    1.55
     previous
    1.45
     recently
    1.24
    previously
    1.21
     recent
    1.11
     предыду
    1.10
     ранее
    1.09
     was
    1.08
     Previously
    1.08
    進んで
    1.06
    Activations Density 0.038%

    No Known Activations