© 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. 8440
    Prev
    Next
    INDEX
    Explanations

    Or, if, want, first

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

    The neuron is chiefly activated by inflected (non‐base) verb forms—i.e. past participles (–ed) and gerunds (–ing).

    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
    ج
    0.80
    ม
    0.77
    غ
    0.77
    д
    0.75
    נו
    0.73
    नई
    0.72
    व
    0.72
    श
    0.71
    ཝ
    0.71
    का
    0.70
    POSITIVE LOGITS
     orchid
    0.92
    ives
    0.86
     orchids
    0.85
    烁
    0.82
     saffron
    0.81
    INESS
    0.80
     somatic
    0.80
     hypertrophy
    0.80
     gauze
    0.80
    arag
    0.79
    Activations Density 0.000%

    No Known Activations