© 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. 87224
    Prev
    Next
    INDEX
    Explanations

    review or recommended

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

    The neuron fires on sequences of digits—i.e. numerals such as years, page counts, issue numbers, prices, ISBN numbers, and other purely numeric tokens.

    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
     blumen
    -0.79
    ҹ
    -0.77
     определение
    -0.76
     culminating
    -0.75
    renia
    -0.74
     zuerst
    -0.73
    ovas
    -0.73
    boston
    -0.73
     bestäm
    -0.72
     (“
    -0.72
    POSITIVE LOGITS
    CHOICE
    1.55
     CHOICE
    1.48
    Choice
    1.33
     Choice
    1.29
    choice
    1.17
     choice
    1.16
    Summing
    1.04
    recommended
    0.99
    choices
    0.97
    Reviewed
    0.96
    Activations Density 0.016%

    No Known Activations