© 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. 34-GEMMASCOPE-RES-131K
    7. 77030
    Prev
    Next
    INDEX
    Explanations

    permanently, permafrost, emanation, permanent

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

    The neuron activates on any token fragment containing “perma” (e.g. in “permanently,” “permafrost,” “perma-waxed”).

    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
     Plural
    -0.82
     Ratna
    -0.80
     Happens
    -0.74
     KAN
    -0.72
     comps
    -0.69
    Kä
    -0.69
    retweet
    -0.68
     movements
    -0.68
     ν
    -0.68
    caus
    -0.68
    POSITIVE LOGITS
     permanence
    0.89
     militare
    0.80
    ence
    0.80
    frost
    0.79
    ently
    0.77
    ense
    0.75
    cznej
    0.75
    ্প
    0.74
    ằm
    0.74
    ental
    0.72
    Activations Density 0.030%

    No Known Activations