© 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. 26353
    Prev
    Next
    INDEX
    Explanations

    measurement context

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

    The neuron strongly activates on alphanumeric tokens that look like identifiers or numeric literals (e.g. file‐path segments, variable names, large numbers).

    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
     aiutare
    -1.06
     العديد
    -0.95
     any
    -0.94
     Mehrheit
    -0.89
     when
    -0.88
     ऐसे
    -0.86
     haría
    -0.86
     dicono
    -0.85
     fornire
    -0.85
     portare
    -0.84
    POSITIVE LOGITS
     amorosa
    1.05
     they
    0.99
     fevereiro
    0.88
    ATUM
    0.88
    поги
    0.86
     остальные
    0.83
    ￠
    0.82
    AUG
    0.82
     prochains
    0.81
     They
    0.81
    Activations Density 0.102%

    No Known Activations