© 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. 128111
    Prev
    Next
    INDEX
    Explanations

    The neuron doesn’t respond to any of the visible tokens—it remains inactive on all input shown.

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

    census counts and subsequent data

    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
     whose
    -1.25
     who
    -1.09
     any
    -1.08
     unique
    -1.04
    ريد
    -1.00
     that
    -0.99
     piaci
    -0.98
     jagung
    -0.98
     izin
    -0.96
    ukunfts
    -0.96
    POSITIVE LOGITS
    舛
    1.00
     вновь
    0.97
     Hinsicht
    0.96
    dagogik
    0.96
     posteriormente
    0.96
    Subscribers
    0.96
     triples
    0.94
     Subsequently
    0.93
     –
    0.91
     оконча
    0.91
    Activations Density 0.000%

    No Known Activations