© 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. 9424
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on occurrences of the verb “look” (e.g. “look,” “look at,” “look forward”).

    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
    у
    1.57
    ра
    1.51
    ﺵ
    1.47
    出血
    1.46
     oscura
    1.43
    ల
    1.43
    не
    1.42
    濆
    1.41
     помощь
    1.38
    ся
    1.35
    POSITIVE LOGITS
    ية
    1.36
    alike
    1.31
    و
    1.27
    ahead
    1.26
    off
    1.23
    ups
    1.23
    粼
    1.20
    ավոր
    1.18
    ian
    1.17
    el
    1.12
    Activations Density 0.100%

    No Known Activations