© 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. 34-GEMMASCOPE-RES-131K
    7. 98950
    Prev
    Next
    INDEX
    Explanations

    encoding base64 images

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

    The neuron detects references to “Base 64” (i.e. the tokens “base,” “6,” and “4”) indicating Base64 encoding.

    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
    lecular
    -0.87
     acompaña
    -0.81
    <0x97>
    -0.79
     bambú
    -0.78
    groet
    -0.76
     WAN
    -0.75
    lastonbury
    -0.75
     удачи
    -0.75
    WAN
    -0.75
    äus
    -0.74
    POSITIVE LOGITS
     encoding
    1.73
     encoded
    1.60
     encode
    1.59
    encode
    1.53
    Encoding
    1.32
    encoding
    1.30
    Encode
    1.30
     Encode
    1.26
    decode
    1.23
     encodes
    1.22
    Activations Density 0.035%

    No Known Activations