© 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. 34-GEMMASCOPE-RES-131K
    7. 116420
    Prev
    Next
    INDEX
    Explanations

    academic or technical identifiers

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

    The neuron detects short journal or DOI identifier tokens (e.g. “pone,” “mec,” “tid,” “pai,” “advs,” “figshare”) that appear in publication citations or URLs.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     both
    -0.86
     much
    -0.83
     over
    -0.82
     Seems
    -0.80
     seemingly
    -0.77
     could
    -0.76
     all
    -0.74
     while
    -0.73
     wide
    -0.73
     rapidly
    -0.73
    POSITIVE LOGITS
    Wenn
    0.89
     DAO
    0.88
     emporter
    0.88
     attirer
    0.88
    THAT
    0.85
    Scott
    0.85
     tilby
    0.85
     trám
    0.85
    hoek
    0.84
    ritis
    0.82
    Activations Density 0.005%

    No Known Activations