© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 12903
    Prev
    Next
    INDEX
    Explanations

    hexadecimal strings

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

    This neuron detects garbled or “mojibake” text sequences (corrupted multi-byte/non-ASCII characters).

    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
    u
    0.57
    it
    0.56
    et
    0.54
    l
    0.54
    y
    0.53
    r
    0.52
    c
    0.49
    t
    0.49
    h
    0.47
    esterday
    0.47
    POSITIVE LOGITS
     Adrián
    0.46
     Álvaro
    0.41
    を含む
    0.39
    に加え
    0.39
    さえ
    0.38
    に対する
    0.38
    <unused357>
    0.38
    DeviceCompliance
    0.38
    စျေး
    0.38
    をはじめ
    0.37
    Activations Density 0.000%

    No Known Activations