© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 382
    Prev
    Next
    INDEX
    Explanations

    also

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

    The main thing this neuron does is detect numeric performance statistics (counts, scores, and related numerical values).

    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
     можуть
    -1.16
     jú
    -1.14
    Zitat
    -1.12
    łóż
    -1.09
     anfangen
    -1.09
     dál
    -1.09
    まずは
    -1.09
    یط
    -1.08
     Hivatalos
    -1.07
     stratégique
    -1.07
    POSITIVE LOGITS
     also
    2.63
     también
    1.52
     همچنین
    1.50
    also
    1.40
     também
    1.37
    Also
    1.37
     также
    1.31
     även
    1.31
     myös
    1.31
     dessutom
    1.27
    Activations Density 0.045%

    No Known Activations