© 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. 128654
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on standalone title- or heading-style tokens (e.g. capitalized or label words used as section names or tags).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    backend words

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     offered
    -1.00
     hesit
    -0.86
     Sailors
    -0.83
     possibilities
    -0.83
    hci
    -0.81
     financed
    -0.81
     phenotypic
    -0.81
     போ
    -0.79
     MRP
    -0.78
     εμπ
    -0.78
    POSITIVE LOGITS
     behind
    1.25
    behind
    1.24
     internal
    1.19
     hidden
    1.17
     backend
    1.15
    backend
    1.11
     backstage
    1.05
     پشت
    1.02
    Behind
    1.00
    ctique
    0.97
    Activations Density 0.153%

    No Known Activations