© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 11808
    Prev
    Next
    INDEX
    Explanations

    guess the letter

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

    The neuron fires strongly on the first significant content word at the start of each new paragraph or section, effectively marking paragraph/section boundaries.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    duplicated or repeated text segments within documents.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    regular past-tense verbs ending in "ed".

    oai_token-act-pair · gpt-5Triggered 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
    house
    0.73
     высо
    0.73
    льной
    0.72
     způsob
    0.70
    сный
    0.70
     bangsa
    0.68
    種類
    0.67
     aswell
    0.67
    ed
    0.67
    ferences
    0.67
    POSITIVE LOGITS
     allegation
    0.82
     데이터를
    0.82
    ВО
    0.79
    ית
    0.79
    हां
    0.74
    ➁
    0.73
    ँग
    0.72
     handbag
    0.72
     considers
    0.72
     প্রত্যা
    0.71
    Activations Density 0.000%

    No Known Activations