© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 70370
    Prev
    Next
    INDEX
    Explanations

    introducing explanation

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

    The neuron activates on modal and conditional cues—words like “if,” “can,” “will,” and “you”—marking instructional or procedural statements (e.g. “if doing X, you can expect Y”).

    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
     of
    -2.27
     in
    -2.16
     this
    -2.09
     to
    -2.03
     because
    -1.87
     on
    -1.79
     these
    -1.79
     at
    -1.76
     there
    -1.75
     or
    -1.73
    POSITIVE LOGITS
    lllllll
    1.73
     verhind
    1.66
     defekt
    1.59
    llll
    1.55
    糁
    1.55
    iamo
    1.55
     ส์
    1.53
    lllll
    1.51
    usions
    1.49
    ᜀ
    1.49
    Activations Density 0.020%

    No Known Activations