© 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. 2726
    Prev
    Next
    INDEX
    Explanations

    lists. parts. such. display

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

    This neuron fires on tokens in procedural, step-by-step instruction text (e.g. the imperative and connective words and punctuation of how-to directions).

    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
     click
    -0.99
     CLICK
    -0.93
     clicked
    -0.88
    鼠标
    -0.83
    ⇑
    -0.82
     ob
    -0.81
    Stef
    -0.80
    クリック
    -0.80
    theless
    -0.79
     murdered
    -0.79
    POSITIVE LOGITS
    人員
    0.89
    Talk
    0.88
    touch
    0.83
    給你
    0.83
     icons
    0.79
     opzioni
    0.79
    Touch
    0.79
    可能會
    0.78
    gls
    0.77
    抬手
    0.77
    Activations Density 0.015%

    No Known Activations