© 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. 34-GEMMASCOPE-RES-131K
    7. 47497
    Prev
    Next
    INDEX
    Explanations

    study trial case

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

    The neuron fires strongly on single-word labels or keywords that head off sections or actions in technical or research texts (e.g. “Study,” “Trial,” “case,” “task,” “modify”).

    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
    puh
    -0.79
     wobei
    -0.79
    acción
    -0.78
     what
    -0.77
     клі
    -0.77
     Unesco
    -0.76
    uebe
    -0.75
     embolism
    -0.75
     Alvarado
    -0.75
     tutu
    -0.72
    POSITIVE LOGITS
    氵
    0.98
     nuevos
    0.93
    <.
    0.88
     trial
    0.85
     triun
    0.81
     recruiting
    0.80
     duration
    0.80
     leur
    0.79
     number
    0.79
     intptr
    0.79
    Activations Density 0.000%

    No Known Activations