© 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. 22-GEMMASCOPE-RES-131K
    7. 38949
    Prev
    Next
    INDEX
    Explanations

    codes and specific terms

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

    The neuron fires on single-word tokens that start with a capital letter followed by lowercase letters (e.g. "Primitives" or "Mh").

    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
     for
    -1.70
     from
    -1.46
     his
    -1.45
     all
    -1.36
     it
    -1.32
     P
    -1.30
     N
    -1.20
    踵
    -1.20
     on
    -1.19
     large
    -1.19
    POSITIVE LOGITS
    küche
    1.43
    Nasc
    1.43
     nettoyer
    1.36
    Intérieur
    1.33
     作り方
    1.32
     チラ
    1.32
     faciliter
    1.31
     schuhe
    1.31
     empêcher
    1.27
     demokra
    1.27
    Activations Density 0.000%

    No Known Activations