© 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. 9112
    Prev
    Next
    INDEX
    Explanations

    grouping and structure

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

    The neuron selectively activates on longer, multi-syllable content words (e.g. complex verbs and past participles).

    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
    단의
    0.78
    으로서
    0.76
     অস্বীক
    0.76
    ambito
    0.73
     কাহারও
    0.73
     attorney
    0.70
    ਅ
    0.69
    ]>
    0.68
     đoán
    0.68
    噌
    0.68
    POSITIVE LOGITS
    ik
    0.96
    od
    0.90
     kuchen
    0.83
    ة
    0.80
    in
    0.79
    at
    0.79
    out
    0.79
     خراب
    0.77
    cigarette
    0.76
    io
    0.75
    Activations Density 0.000%

    No Known Activations