© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67169
    Prev
    Next
    INDEX
    Explanations

    asking questions and making inquiries

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

    The neuron detects social‐proof cues—words signaling reported speech or testimonial framing (e.g. “asked,” “people,” “say,” “friends,” 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_22/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
     also
    -0.93
     might
    -0.92
     никого
    -0.89
     stagione
    -0.85
     could
    -0.85
    erencias
    -0.85
     opgenomen
    -0.82
    -0.82
     didel
    -0.82
     probably
    -0.81
    POSITIVE LOGITS
     my
    1.25
     aceites
    1.04
     pregunt
    1.02
     inquiries
    1.02
     acquaintances
    1.02
    either
    1.00
     asking
    0.99
     contacts
    0.98
     friends
    0.98
     questions
    0.98
    Activations Density 0.033%

    No Known Activations