© 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. 10-GEMMASCOPE-RES-131K
    7. 9455
    Prev
    Next
    INDEX
    Explanations

    what they do

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

    The neuron strongly lights up on short slogan-style statements that pair personal pronouns (I, you, we, they) with action or purpose verbs to express beliefs, reasons, or motivations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     of
    -2.53
     homicidio
    -1.62
     our
    -1.59
     unserer
    -1.47
     delito
    -1.40
    вся
    -1.39
     comenzaron
    -1.34
     ακόμα
    -1.23
     his
    -1.20
     nossas
    -1.20
    POSITIVE LOGITS
     it
    2.00
    ##
    1.57
     hôte
    1.40
    --
    1.39
    ֶּ
    1.30
     осмо
    1.29
     Outils
    1.28
    するもの
    1.27
     beslu
    1.27
     Ө
    1.27
    Activations Density 0.059%

    No Known Activations