© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 7556
    Prev
    Next
    INDEX
    Explanations

    file not found

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

    The neuron fires on terms signaling error or failure messages (e.g. “files not found,” “timeout,” “unsupported format”).

    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
    は
    0.66
    Là
    0.62
    AndTime
    0.62
    是
    0.62
    nbsp
    0.59
    和
    0.58
    Validate
    0.57
    NotFound
    0.57
    unread
    0.57
    Communication
    0.56
    POSITIVE LOGITS
    !',
    1.33
    !');
    1.27
    !:
    1.23
    !",
    1.16
    !';
    1.16
    !");
    1.13
    !')
    1.13
    !’
    1.11
    !,
    1.09
    !'
    1.09
    Activations Density 0.048%

    No Known Activations