© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25042
    Prev
    Next
    INDEX
    Explanations

    data file list array

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

    The neuron flags code‐style tokens (operators, punctuation, identifiers, numeric constants, etc.), i.e. bits of source code rather than natural‐language words.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     รี
    -0.88
    表彰
    -0.81
    czeń
    -0.81
    Expend
    -0.80
     обслужи
    -0.80
     retir
    -0.79
    ENED
    -0.78
    鈾
    -0.78
    alucía
    -0.78
    кая
    -0.76
    POSITIVE LOGITS
     to
    1.07
     '../
    0.99
     submitted
    0.97
     transfer
    0.95
     prepared
    0.92
    input
    0.89
    len
    0.87
     replacement
    0.87
     './
    0.85
     installation
    0.85
    Activations Density 0.117%

    No Known Activations