© 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. 5673
    Prev
    Next
    INDEX
    Explanations

    colors, movement, riddles, law, math, and journeys

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

    The neuron fires strongly on words that begin new sentences or clauses, i.e. sentence-initial 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
    伉
    0.80
     j
    0.76
     وعند
    0.75
     f
    0.72
     match
    0.71
     وعلى
    0.69
     premise
    0.68
     key
    0.67
     manoeuvre
    0.66
     ent
    0.65
    POSITIVE LOGITS
    りん
    0.91
     chhoti
    0.89
    ěji
    0.85
    ｺ
    0.82
    ラクマ
    0.80
    ק
    0.79
     kaisi
    0.79
     inheriting
    0.78
    ўкі
    0.77
    γκ
    0.77
    Activations Density 0.001%

    No Known Activations