© 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. 22-GEMMASCOPE-RES-131K
    7. 15122
    Prev
    Next
    INDEX
    Explanations

    Podcasting

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

    The neuron fires on runs of non‐alphanumeric punctuation or symbol characters (e.g. clusters of ~, ., :, <, 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
     what
    -1.58
     сожа
    -1.52
    jadikan
    -1.51
     moverse
    -1.47
    Generally
    -1.47
    During
    -1.44
     srd
    -1.44
    Usually
    -1.43
     it
    -1.42
     by
    -1.42
    POSITIVE LOGITS
    !!!!!!!
    1.70
     alcuni
    1.68
    !!!!!!
    1.66
     !!!!
    1.62
     alcune
    1.59
     tumultuous
    1.55
     !!!!!
    1.55
     opulent
    1.47
    或者是
    1.44
    !!!!!!!!!
    1.38
    Activations Density 0.002%

    No Known Activations