© 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. 16162
    Prev
    Next
    INDEX
    Explanations

    events or outcomes

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

    The neuron activates on numeric tokens (years, dates, measurements—any tokens containing digits).

    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
     you
    -1.16
    で開催
    -1.15
     !!!!
    -1.13
     recently
    -1.10
     your
    -1.08
    Opere
    -1.06
     our
    -1.05
    dtp
    -1.04
     we
    -1.02
    very
    -1.02
    POSITIVE LOGITS
     begins
    1.53
    ῖς
    1.48
     after
    1.45
     begin
    1.22
    اقرأ
    1.20
    Dış
    1.18
     unequal
    1.16
    𝟐
    1.16
     začet
    1.16
     began
    1.14
    Activations Density 0.016%

    No Known Activations