© 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. 22-GEMMASCOPE-RES-131K
    7. 129535
    Prev
    Next
    INDEX
    Explanations

    The neuron predominantly lights up on contentful action words—i.e. verbs that describe doing something.

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

    put into, then place

    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
    imine
    -0.90
    posium
    -0.84
     on
    -0.83
    надцать
    -0.82
     や
    -0.80
     الو
    -0.79
     tuvo
    -0.79
     khiến
    -0.79
     médical
    -0.78
     jūs
    -0.78
    POSITIVE LOGITS
     then
    1.65
    放到
    1.09
    Then
    1.01
     затем
    1.00
    然后
    1.00
    แล้ว
    0.96
     from
    0.94
    then
    0.93
    进行
    0.92
     עבור
    0.90
    Activations Density 0.128%

    No Known Activations