© 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. 92358
    Prev
    Next
    INDEX
    Explanations

    Neurohormones

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

    computer programming code structures, specifically Objective-C class declarations.

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

    The neuron is strongly activated by multi‐digit numeric tokens (e.g. long sequences of digits) in the text.

    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
     Golfo
    -0.83
    ┑
    -0.77
     bomba
    -0.76
    TSP
    -0.71
     Gulf
    -0.71
    adă
    -0.71
    밝
    -0.70
    izzati
    -0.70
     cuán
    -0.70
    acos
    -0.69
    POSITIVE LOGITS
     California
    1.10
     CALIFORNIA
    0.93
     california
    0.90
     SB
    0.85
    SB
    0.85
    California
    0.83
     Delano
    0.82
     snowboard
    0.81
    rat
    0.79
     snowboarding
    0.79
    Activations Density 0.018%

    No Known Activations