© 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. 90172
    Prev
    Next
    INDEX
    Explanations

    import statements and biological terms

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

    numerical years (especially those around 2000) and code-related text elements.

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

    The neuron fires on numeric tokens—especially multi‐digit numbers like ages, years, measurements, or other standalone numerals.

    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
     Ministry
    -0.79
    ʒ
    -0.76
     собак
    -0.75
     eink
    -0.73
     jig
    -0.73
     Moody
    -0.72
     Universidad
    -0.71
    cooldown
    -0.71
     thigh
    -0.70
    Philosophy
    -0.70
    POSITIVE LOGITS
     сохра
    0.89
     lapsed
    0.79
     disparate
    0.76
    ონ
    0.73
    psons
    0.71
     gradualmente
    0.71
     heathen
    0.70
    ഛ
    0.69
     ranting
    0.69
    essere
    0.69
    Activations Density 0.001%

    No Known Activations