© 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. 57950
    Prev
    Next
    INDEX
    Explanations

    Microscopic Certification

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

    encoded data formats like RSA keys, base64 strings, and API keys.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates on PEM‐style RSA private key blocks, i.e. the “-----BEGIN RSA PRIVATE KEY-----” header/footer and the long Base64‐encoded key content.

    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
     its
    -0.85
     At
    -0.85
     on
    -0.84
     after
    -0.84
     Also
    -0.82
     with
    -0.80
     because
    -0.80
     Knowing
    -0.79
     found
    -0.79
     and
    -0.78
    POSITIVE LOGITS
     vodou
    1.00
    ﹪
    0.99
     lác
    0.98
    ilirsiniz
    0.98
     سامسونگ
    0.95
     totalt
    0.94
    klare
    0.94
     configurações
    0.93
     sépar
    0.93
     convenable
    0.93
    Activations Density 0.007%

    No Known Activations