© 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-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 8152
    Prev
    Next
    INDEX
    Explanations

    uppercase and lowercase

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

    This neuron detects references to letter‐case conventions—e.g. mentions of uppercase vs. lowercase letters and case‐sensitivity instructions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    words or phrases related to typography and text formatting, particularly instructions about capitalization and proper text presentation.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    mentions of capitalization and case sensitivity, along with broader orthographic guidance about letters, spelling, punctuation, and typing usage.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ವ
    0.55
     SeekBar
    0.54
    纳米
    0.54
     Nan
    0.52
    वित
    0.52
    వల
    0.52
     上海
    0.52
    uaje
    0.50
    糇
    0.50
     Shanghai
    0.50
    POSITIVE LOGITS
     uppercase
    2.48
     capital
    2.44
    uppercase
    2.42
     capitalized
    2.41
    capital
    2.36
     lowercase
    2.33
     capitals
    2.31
     capitalize
    2.28
    Capital
    2.28
     CAPITAL
    2.27
    Activations Density 0.075%

    No Known Activations