© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 99892
    Prev
    Next
    INDEX
    Explanations

    words starting with 'E' and vowel, or containing 'ox'/'epo'; 'erectile funct'

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

    proper nouns and brand names, particularly at the beginning of words.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron is triggered by three-letter sequences of the form “E?o” (e.g. Eno, Elo, Ero, Epo) regardless of the surrounding context.

    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
    OSPH
    -0.78
    בות
    -0.77
    willing
    -0.76
    路径
    -0.75
     skier
    -0.71
    REQUIRES
    -0.71
    employeeId
    -0.70
    の位置
    -0.70
    Obtener
    -0.69
    봅
    -0.68
    POSITIVE LOGITS
     Correlation
    0.80
     Repeat
    0.78
     abandoned
    0.77
     Abandoned
    0.76
     Norge
    0.76
     whence
    0.75
     Medal
    0.73
    ALT
    0.72
    lede
    0.70
     巨
    0.69
    Activations Density 0.027%

    No Known Activations