© 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. 25672
    Prev
    Next
    INDEX
    Explanations

    clocking, January, number fields

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

    This neuron fires on tokens that appear in live chat or IRC–style dialogue, such as the “live:” prefix, emoticons (“;)”), and colloquial interjections (“whats that?”).

    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
     encontraban
    -1.16
     retten
    -1.15
     encontraba
    -1.15
    rrrrr
    -1.13
     aplique
    -1.11
    livejournal
    -1.11
     condado
    -1.09
     pertanian
    -1.07
     zü
    -1.05
     disfrutando
    -1.05
    POSITIVE LOGITS
     …
    1.44
     exult
    1.29
     obstin
    1.24
    "—
    1.23
     πρ
    1.23
     impot
    1.18
     intoler
    1.16
     immor
    1.13
     fathoms
    1.09
     directs
    1.08
    Activations Density 0.046%

    No Known Activations