© 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. 2294
    Prev
    Next
    INDEX
    Explanations

    still followed by state or process

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

    The neuron fires on “still” and other markers of ongoing or continuing states (e.g. “still wanting,” “still my favorite,” “still have…”).

    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
     beetje
    0.96
     dignitaries
    0.95
     artık
    0.92
    apan
    0.91
    yemek
    0.89
    arctan
    0.89
     convertirse
    0.89
    蹶
    0.87
     AppMethodBeat
    0.87
     tornar
    0.86
    POSITIVE LOGITS
    คง
    0.96
    ไง
    0.96
    birth
    0.94
    m
    0.92
     остается
    0.91
     plenty
    0.89
     ድረስ
    0.88
     남아
    0.88
    ্ড
    0.86
     불구하고
    0.86
    Activations Density 0.276%

    No Known Activations