© 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. 31-GEMMASCOPE-2-RES-16K
    7. 10166
    Prev
    Next
    INDEX
    Explanations

    python test runner

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

    The neuron is specialized to detect long runs or clusters of question-mark and similar punctuation sequences.

    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
    ◃
    0.58
    韫
    0.55
    Encoder
    0.52
    impse
    0.52
    FullPath
    0.52
    reonine
    0.52
    Conversion
    0.51
     Verarbeitung
    0.51
    মাণ
    0.50
    ucumber
    0.50
    POSITIVE LOGITS
     ok
    0.67
     $(\
    0.67
    ok
    0.66
     loy
    0.64
     )),
    0.63
    ))$,
    0.63
     todo
    0.61
     KYC
    0.61
     Vad
    0.60
     ))
    0.59
    Activations Density 0.020%

    No Known Activations