© 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. 13512
    Prev
    Next
    INDEX
    Explanations

    try again later

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

    The main thing this neuron does is detect error or failure messages that apologize and ask the user to “please try again” (e.g. service-unavailable or request-failed prompts).

    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_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
    癣
    0.73
    搜索
    0.68
    Debugging
    0.68
    Pokud
    0.68
    Archae
    0.63
    検索
    0.63
    蹰
    0.61
    停车
    0.61
     Porém
    0.61
    grayscale
    0.60
    POSITIVE LOGITS
     loneliness
    0.63
     triumph
    0.57
    watcher
    0.57
     canola
    0.56
     lím
    0.56
     joy
    0.55
     edible
    0.55
     Calab
    0.55
     surface
    0.54
    entin
    0.54
    Activations Density 0.001%

    No Known Activations