© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 83689
    Prev
    Next
    INDEX
    Explanations

    hexadecimal identifiers

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

    The neuron is tuned to spot long sequences of hexadecimal characters (e.g. checksums, hashes, GUID fragments).

    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
     tutela
    -0.94
     tougher
    -0.92
    ギフト
    -0.90
     after
    -0.90
     voidaan
    -0.88
     proposito
    -0.84
     terribly
    -0.83
     フラ
    -0.82
     especially
    -0.81
     fric
    -0.81
    POSITIVE LOGITS
     reconhe
    1.03
    €™
    1.01
     visitantes
    1.00
     joyería
    0.98
     Sz
    0.96
    gemene
    0.96
     rekom
    0.96
     mecánica
    0.95
     recién
    0.95
     jabón
    0.95
    Activations Density 0.015%

    No Known Activations