© 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. 25481
    Prev
    Next
    INDEX
    Explanations

    their relationships

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

    The neuron responds to highly frequent English function words (like “and,” “the,” “or,” “what,” etc.).

    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
     matériau
    -0.96
     preferred
    -0.93
     erfordert
    -0.89
     diversity
    -0.88
    Consigli
    -0.86
    Catatan
    -0.84
     available
    -0.84
    Sito
    -0.84
    Preise
    -0.84
     constituency
    -0.84
    POSITIVE LOGITS
     reflexiones
    0.99
     their
    0.94
     pensamientos
    0.92
     manifestaciones
    0.90
    associated
    0.88
    ͆
    0.86
     eenig
    0.84
    ({\
    0.84
    ogues
    0.84
    belts
    0.83
    Activations Density 0.047%

    No Known Activations