© 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. 15641
    Prev
    Next
    INDEX
    Explanations

    at, with, by the

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

    This neuron activates on words expressing personal emotional reactions or sentiments (e.g., pleased, amazed, impressed, shocked, disappointed).

    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
     Probably
    -1.80
     Many
    -1.79
     Perhaps
    -1.75
     Actually
    -1.74
     Sometimes
    -1.74
     Some
    -1.71
     Those
    -1.71
     Makes
    -1.69
     before
    -1.68
     Here
    -1.68
    POSITIVE LOGITS
     the
    2.94
     how
    2.02
    橚
    1.73
    鳚
    1.45
     impecable
    1.41
    盉
    1.38
    having
    1.37
     verhind
    1.36
     pastillas
    1.34
     Minggu
    1.34
    Activations Density 0.027%

    No Known Activations