© 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. 33121
    Prev
    Next
    INDEX
    Explanations

    acceleration (m/s^2)

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

    The neuron consistently fires on tokens that consist of or contain numeric digits (i.e. number literals).

    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
     which
    -1.15
     where
    -1.12
     if
    -1.08
     when
    -1.03
     this
    -1.02
    店員
    -0.95
    ürm
    -0.89
     for
    -0.85
     mussten
    -0.85
     their
    -0.84
    POSITIVE LOGITS
    actéristi
    1.23
     into
    1.20
     soooo
    1.17
    (")");
    1.13
     sooo
    1.13
     ravaged
    1.12
     mediocre
    1.06
     delightfully
    1.04
     usur
    1.03
     fervent
    1.02
    Activations Density 0.007%

    No Known Activations