© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 30875
    Prev
    Next
    INDEX
    Explanations

    HTTP status 200

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

    Explanation of neuron 4 behavior: the main thing this neuron does is find numeric tokens (especially HTTP response/status code digits).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    4
    -1.02
    ৪
    -0.93
     stderr
    -0.90
    слен
    -0.86
    ４
    -0.84
     czter
    -0.82
    öten
    -0.81
    Fehler
    -0.81
    ۴
    -0.80
     negative
    -0.79
    POSITIVE LOGITS
     OK
    2.02
    OK
    1.78
     ok
    1.73
     okay
    1.73
    Ok
    1.65
    ok
    1.64
     Ok
    1.62
     Okay
    1.51
    Okay
    1.51
    StatusOK
    1.33
    Activations Density 0.007%

    No Known Activations