© 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. 117288
    Prev
    Next
    INDEX
    Explanations

    I can think of

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

    This neuron fires on first-person self-reference and related modal/reflective verbs (e.g. “I,” “can,” “could,” “think,” 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
     AREAS
    -0.74
    astră
    -0.74
    概述
    -0.73
    маны
    -0.73
    Resizable
    -0.72
    эн
    -0.71
    anganese
    -0.71
    Wikis
    -0.71
    olated
    -0.70
    我希望
    -0.70
    POSITIVE LOGITS
     think
    7.38
     thinking
    6.56
    think
    5.66
     Think
    5.66
    Think
    5.63
     thinks
    5.00
     THINK
    4.78
     pensar
    4.72
    thinking
    4.66
     Thinking
    4.56
    Activations Density 0.151%

    No Known Activations