© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 1902
    Prev
    Next
    INDEX
    Explanations

    look then

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

    The neuron is keyed to discourse markers and transition cues (e.g. “whether,” “I mean,” “maybe,” “which,” “so,” “but,” etc.) that signal elaboration, contrast, or examples.

    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
     após
    0.73
    했으며
    0.71
     Jika
    0.70
     있으며
    0.68
     incap
    0.64
     пр
    0.64
     сможет
    0.64
    ()/
    0.64
     jika
    0.63
     способен
    0.63
    POSITIVE LOGITS
     इंपॉर्टेंट
    0.81
     ờ
    0.79
     बेसिकली
    0.78
    ustos
    0.78
     äh
    0.73
     ähm
    0.73
    unker
    0.73
    zeugt
    0.73
    itter
    0.73
    uzi
    0.73
    Activations Density 0.081%

    No Known Activations