© 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. 32910
    Prev
    Next
    INDEX
    Explanations

    custom code or specific queries

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

    This neuron detects line-initial capitalized words (the first word at the start of a new line or sentence).

    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
    ificação
    -0.85
    kurtka
    -0.84
     potenci
    -0.84
    menjadi
    -0.83
    请问
    -0.82
     annonce
    -0.79
    ژاد
    -0.77
    ⿵
    -0.77
    ⿺
    -0.77
     egip
    -0.77
    POSITIVE LOGITS
     Fli
    0.84
    ínica
    0.80
    variates
    0.80
    Wildcard
    0.79
    aternal
    0.77
     Forder
    0.77
    otus
    0.77
     بر
    0.77
     Weird
    0.76
    х
    0.76
    Activations Density 0.000%

    No Known Activations