© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 1719
    Prev
    Next
    INDEX
    Explanations

    clearer auscultation

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

    The neuron responds to words that denote loud vocal outbursts or exclamations (e.g. shouting, yelling, screaming, crying out).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     phẳng
    0.67
     erweitert
    0.61
     Erweiter
    0.56
     densité
    0.55
    🎞
    0.54
     porosity
    0.53
    ක්ෂ
    0.53
     केसा
    0.53
     স্থিত
    0.53
    撣
    0.53
    POSITIVE LOGITS
     louder
    2.05
     loud
    2.02
     loudly
    1.91
     loudest
    1.90
    loud
    1.87
     shouting
    1.80
     shouted
    1.77
     shouts
    1.74
     shout
    1.64
     yelling
    1.61
    Activations Density 0.551%

    No Known Activations