© 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. 65310
    Prev
    Next
    INDEX
    Explanations

    time periods and places

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

    The neuron consistently activates on numerical tokens—digits, ages, time spans, and other number expressions—in the text.

    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
    !
    -1.47
     our
    -1.36
     what
    -1.36
     so
    -1.31
     τὴν
    -1.29
     even
    -1.27
     perhaps
    -1.27
     (
    -1.27
     τὸν
    -1.26
     really
    -1.24
    POSITIVE LOGITS
    atives
    1.28
    ְּ
    1.28
    ּוֹ
    1.25
    ᾿
    1.24
     ultime
    1.22
    osos
    1.21
     FOOTBALL
    1.20
    ֶּ
    1.20
    ֵּ
    1.20
     habitants
    1.18
    Activations Density 0.194%

    No Known Activations