© 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. 80887
    Prev
    Next
    INDEX
    Explanations

    Unicode text processing

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

    The neuron strongly activates on numeric tokens (digits, version numbers, dates) in the code.

    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
    Estudios
    -0.69
    resentation
    -0.69
    大声
    -0.65
    Eliminar
    -0.64
    uerdo
    -0.64
     hoodie
    -0.64
     hoodies
    -0.63
     planks
    -0.63
     forwards
    -0.63
     toán
    -0.63
    POSITIVE LOGITS
    ари
    0.82
    キス
    0.72
     Hra
    0.71
    ENGL
    0.71
    oudre
    0.69
    zeczy
    0.68
     Kitt
    0.66
     Drogen
    0.66
     podob
    0.66
     ideas
    0.65
    Activations Density 0.039%

    No Known Activations