© 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. 10-GEMMASCOPE-RES-131K
    7. 44874
    Prev
    Next
    INDEX
    Explanations

    bluff, poker, bullshit, chicken

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

    The neuron fires on words that convey bold posturing or deceptive bravado (e.g. bluffing, bluster, moxie, game of chicken, poker, bullshit).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    쨩
    -1.75
    注意事
    -1.71
    !’
    -1.59
    -1.56
    !'
    -1.55
    (‘
    -1.54
    玙
    -1.52
    哢
    -1.52
     Their
    -1.50
     programma
    -1.49
    POSITIVE LOGITS
     komplette
    1.48
    $,
    1.44
     incite
    1.39
    try
    1.38
     glam
    1.36
     своей
    1.35
     pleas
    1.35
     -
    1.35
     también
    1.32
     manic
    1.32
    Activations Density 0.018%

    No Known Activations