© 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. 31-GEMMASCOPE-2-RES-16K
    7. 6213
    Prev
    Next
    INDEX
    Explanations

    important notes and disclaimers

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

    The neuron responds to emphasis headers—especially the word “Important” (and closely related “information”) used to flag key notices or sections.

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

    the word "important" in formal or instructional text.

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

    statements that emphasize significance or call out key information, such as notices or warnings indicating something is important.

    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_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
     certamente
    0.78
     Freder
    0.68
     certainly
    0.62
     तकलीफ
    0.62
    Certainly
    0.62
    Cit
    0.61
     हालत
    0.61
     disgr
    0.61
     Certainly
    0.60
    رہ
    0.60
    POSITIVE LOGITS
    IMPORTANT
    0.89
     importants
    0.87
    important
    0.82
     ramifications
    0.82
     중요한
    0.81
     IMPORTANT
    0.81
     помнить
    0.80
     vigt
    0.80
     Important
    0.79
     중요
    0.79
    Activations Density 0.128%

    No Known Activations