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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 92736
    Prev
    Next
    INDEX
    Explanations

    long time roles and affiliations

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

    This neuron activates on occurrences of the phrase “long time” (including hyphenated or concatenated forms like “long-time” or “longtime”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    それと
    -0.97
     newly
    -0.97
     начинают
    -0.96
    newly
    -0.93
     indifer
    -0.93
    年轻
    -0.93
     LIABLE
    -0.88
     ludzie
    -0.87
     νέ
    -0.86
    ícias
    -0.85
    POSITIVE LOGITS
     supporter
    1.44
     contributor
    1.27
     member
    1.25
     supporters
    1.25
     fixture
    1.23
     friend
    1.22
     collaborator
    1.20
     fixtures
    1.13
     employee
    1.10
     servant
    1.05
    Activations Density 0.030%

    No Known Activations