© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13378
    Prev
    Next
    INDEX
    Explanations

    date and time qualifiers

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

    The neuron strongly lights up on tokens that are rare or out‐of‐distribution (proper names, foreign words, odd or unusually formatted terms), i.e. it detects uncommon/low‐frequency tokens.

    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_24_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
    ت
    1.02
    u
    0.92
    ir
    0.89
    ur
    0.87
    ص
    0.86
    م
    0.85
    с
    0.84
    ç
    0.82
    čke
    0.82
    ต
    0.82
    POSITIVE LOGITS
     CALCUL
    0.84
     Gleich
    0.79
     решению
    0.79
     июня
    0.78
     математи
    0.76
     benutzt
    0.76
    लेषण
    0.76
     inhomogeneities
    0.75
     qubits
    0.74
     campers
    0.73
    Activations Density 0.000%

    No Known Activations