© 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-3-1B
    5. 7-GEMMASCOPE-2-RES-16K
    7. 943
    Prev
    Next
    INDEX
    Explanations

    foreign words

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

    the neuron detects prominent section or category labels—short, standalone nouns that mark headings, product categories, or list items.

    oai_token-act-pair · gpt-5-miniTriggered by @smilgaveronika

    plural nouns ending with “s,” especially category or list labels.

    oai_token-act-pair · gpt-5Triggered by @smilgaveronika
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-1b-pt/resid_post/layer_7_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
    то
    2.13
    𝘢
    2.13
    tted
    2.09
    았
    2.07
    typal
    2.07
    transmitted
    2.05
     juice
    2.03
    𝘴
    1.99
    tie
    1.96
    tion
    1.93
    POSITIVE LOGITS
    heets
    3.97
    heet
    3.66
    ervices
    3.52
    olutions
    3.52
    creen
    3.39
    chool
    3.33
    mith
    3.15
    creens
    3.12
    wamy
    3.09
    pecific
    3.08
    Activations Density 1.623%

    No Known Activations