© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 107568
    Prev
    Next
    INDEX
    Explanations

    tickle, face, dad

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

    The neuron is keyed to bits of text about humor and laughter (e.g. laugh/funny, “makes faces,” tickled/tickling, question-style “What…?”).

    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
    umab
    -0.88
    ctez
    -0.83
    וּ
    -0.82
     BUF
    -0.80
    ською
    -0.79
    stwie
    -0.78
    crata
    -0.77
    idity
    -0.76
    kre
    -0.75
    jeta
    -0.75
    POSITIVE LOGITS
     tickle
    3.73
     tick
    3.34
     Tick
    2.77
    tick
    2.48
    Tick
    2.47
    TICK
    1.99
     TICK
    1.87
     ticks
    1.34
     itchy
    1.29
    痒
    1.27
    Activations Density 0.026%

    No Known Activations