© 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. 31-GEMMASCOPE-2-RES-16K
    7. 471
    Prev
    Next
    INDEX
    Explanations

    unusual phrasing

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

    The neuron activates strongly on common capitalized words that begin sentences (e.g. “The,” “But,” “My”), effectively detecting sentence‐initial capitalization.

    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
     Nucleaires
    0.27
     nuovamente
    0.27
     ৃ
    0.26
     plötzlich
    0.26
     கூறி
    0.25
     Polizei
    0.25
     enraged
    0.25
     இருவரும்
    0.25
     policía
    0.24
     ///<
    0.24
    POSITIVE LOGITS
     reason
    0.52
     downside
    0.51
     greatest
    0.46
     biggest
    0.46
    理由は
    0.46
     drawback
    0.46
    之所以
    0.45
     이유는
    0.42
     advantage
    0.41
     সবচেয়ে
    0.40
    Activations Density 0.989%

    No Known Activations