© 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. 57583
    Prev
    Next
    INDEX
    Explanations

    terror and Terry names

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

    The neuron detects occurrences of the substring “terror” in words (e.g. “terror,” “terrorist,” “terroriser,” etc.).

    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
    登場
    -0.83
    Lastly
    -0.83
    tamientos
    -0.80
     Lastly
    -0.79
     جای
    -0.78
    ᕼ
    -0.78
    uş
    -0.77
    jira
    -0.77
     Witches
    -0.77
     SDM
    -0.76
    POSITIVE LOGITS
     TERR
    1.15
    rists
    0.96
    ist
    0.95
    terror
    0.90
    cotta
    0.90
    istat
    0.89
    itorial
    0.88
     Terri
    0.88
    aniline
    0.85
     terror
    0.85
    Activations Density 0.012%

    No Known Activations