© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 85913
    Prev
    Next
    INDEX
    Explanations

    Analysis of, How Far, Multistrain

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

    The neuron fires on long, uninterrupted alphanumeric strings typical of Base64‐encoded cryptographic blobs (e.g. RSA/DH key blocks).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    🌸
    -2.58
     if
    -2.45
    🌟
    -2.38
    W
    -2.33
     after
    -2.30
     Putting
    -2.25
     Quite
    -2.20
     Having
    -2.19
     Just
    -2.17
    Y
    -2.17
    POSITIVE LOGITS
     delightfully
    2.69
    despite
    2.69
    囖
    2.69
    albeit
    2.67
    mbgg
    2.67
    厛
    2.67
     cristianos
    2.66
     VERY
    2.58
     ridiculously
    2.55
     gigantic
    2.53
    Activations Density 0.018%

    No Known Activations