© 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. 4071
    Prev
    Next
    INDEX
    Explanations

    code or technical terms

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

    This neuron fires almost exclusively on the first token of a new text block or “section” (i.e. it marks the start of each segment).

    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
    रत
    0.82
     mankind
    0.79
    ব
    0.73
    즘
    0.73
    淼
    0.72
    сть
    0.71
    сред
    0.71
    ৃতি
    0.70
    蹶
    0.69
     worse
    0.69
    POSITIVE LOGITS
     PHOT
    0.80
    decor
    0.80
     ambos
    0.79
    orderby
    0.79
     ambas
    0.79
    整合
    0.78
     ताब
    0.78
    மல
    0.77
    ARG
    0.77
     tolu
    0.77
    Activations Density 0.000%

    No Known Activations