© 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. 24-GEMMASCOPE-2-RES-16K
    4. 8693
    Prev
    Next
    INDEX
    Explanations

    This neuron detects large decimal numbers (floating-point numerical values) in the text.

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

    contractions with "s" (like "it's" or "what's") or digit sequences in the middle of text.

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

    forms of the verb “to be,” especially contracted and past-tense usages used as copulas or auxiliaries.

    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_24_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
    هُ
    0.77
     Radiological
    0.76
    hésitez
    0.75
    حول
    0.74
     Tất
    0.74
    い
    0.73
    лис
    0.73
    م
    0.73
     год
    0.72
    BLACKLIST
    0.72
    POSITIVE LOGITS
     ventures
    0.78
     관계
    0.77
     domestic
    0.76
     marriage
    0.76
     men
    0.74
     venture
    0.74
     manages
    0.74
     partner
    0.73
     vehicle
    0.72
     danger
    0.72
    Activations Density 0.000%

    No Known Activations