© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7174
    Prev
    Next
    INDEX
    Explanations

    greeting, list or code content

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

    This neuron fires strongly on capitalized tokens—especially proper names or words at the start of a sentence.

    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
    Calories
    0.51
     व
    0.50
     merr
    0.50
    erosion
    0.48
    Injection
    0.48
    Scul
    0.47
    Chevron
    0.47
    JOUR
    0.47
    Գ
    0.47
    Scan
    0.47
    POSITIVE LOGITS
    льные
    0.68
    ട്ടുള്ള
    0.67
     خپل
    0.64
     Qualquer
    0.64
    тые
    0.64
     newMovie
    0.64
     StandardScaler
    0.63
    াতাড়ি
    0.62
     topLeft
    0.62
     Бушлай
    0.61
    Activations Density 3.414%

    No Known Activations