© 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. 57435
    Prev
    Next
    INDEX
    Explanations

    deceit, deception, fraud, trickery

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

    The neuron fires on numerical or quantitative information—digits, monetary amounts, counts, and related tokens.

    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
    黠
    -0.74
     nút
    -0.74
    fatal
    -0.74
    btn
    -0.71
    Conflicts
    -0.71
    вид
    -0.71
     inappropriate
    -0.71
    Kok
    -0.70
    Ports
    -0.68
    ionate
    -0.68
    POSITIVE LOGITS
     cheated
    2.61
     cheat
    2.47
     deceive
    2.44
     deception
    2.41
     deceived
    2.31
     cheating
    2.27
     deceiving
    2.14
    cheat
    2.06
    Cheat
    1.98
     cheats
    1.96
    Activations Density 0.029%

    No Known Activations