© 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. 9717
    Prev
    Next
    INDEX
    Explanations

    as a purpose or function

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

    The neuron responds strongly to common English function words (short prepositions and articles) that introduce phrases or list entries.

    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
     particolarmente
    -1.00
     perceber
    -0.99
    няют
    -0.98
     peggio
    -0.97
    我真的
    -0.94
     meravigli
    -0.94
    我才
    -0.93
     their
    -0.93
     altrimenti
    -0.92
     époux
    -0.91
    POSITIVE LOGITS
    羸
    1.06
    ريقيا
    1.04
    ulink
    1.04
    ʼn
    1.02
    ierung
    0.97
    ))->
    0.97
    病例
    0.97
    }):
    0.95
    äude
    0.95
    quième
    0.94
    Activations Density 0.000%

    No Known Activations