© 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. 31-GEMMASCOPE-2-RES-16K
    7. 7843
    Prev
    Next
    INDEX
    Explanations

    asking to do something

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

    The neuron detects non-Latin script tokens, especially Japanese and Korean characters (e.g. Hangul and Kanji/Kana).

    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
     역시
    1.05
    을
    1.02
    를
    1.00
    에
    0.99
    에서
    0.96
    의
    0.96
    에게
    0.89
    들을
    0.88
    가
    0.88
    저
    0.86
    POSITIVE LOGITS
    のだろう
    0.98
    വന്‍
    0.95
    こともある
    0.88
    のではなく
    0.85
    𝒶
    0.84
    ので
    0.84
    んだろう
    0.82
    のに
    0.80
    のです
    0.80
    ための
    0.80
    Activations Density 0.005%

    No Known Activations