© 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. 24-GEMMASCOPE-2-RES-16K
    7. 9610
    Prev
    Next
    INDEX
    Explanations

    acknowledgement of correctness

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

    The neuron strongly detects phrases where the writer expresses agreement or approval (e.g., “I agree,” “you’re right,” “indeed”).

    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
    science
    0.66
    fiction
    0.63
    '
    0.63
    to
    0.63
    f
    0.62
    rumah
    0.61
    localhost
    0.61
    feet
    0.60
     phoned
    0.59
    subscribed
    0.59
    POSITIVE LOGITS
     aumentare
    0.68
     augmenter
    0.67
    ل
    0.62
     remarquer
    0.61
    性が
    0.60
     danneggi
    0.59
     méth
    0.59
     épa
    0.59
     étro
    0.59
     worsening
    0.58
    Activations Density 0.025%

    No Known Activations