© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 33217
    Prev
    Next
    INDEX
    Explanations

    start of code or descriptions

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

    The neuron fires on single‐character tokens (digits or punctuation like dots and parentheses) typically found in code or regex snippets.

    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
     even
    -1.04
     while
    -0.93
     with
    -0.91
     on
    -0.90
     what
    -0.89
     and
    -0.85
     almost
    -0.84
     ${
    -0.82
     arguably
    -0.81
     all
    -0.79
    POSITIVE LOGITS
     alpes
    0.99
     canale
    0.98
    discussed
    0.95
    weird
    0.94
    rased
    0.94
    Curios
    0.92
     Bibl
    0.91
     terminer
    0.90
     CRES
    0.90
     constamment
    0.90
    Activations Density 0.000%

    No Known Activations