© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 10413
    Prev
    Next
    INDEX
    Explanations

    accelerating rate

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

    The neuron fires on occurrences of the substring “Acceler” (e.g. in words like “Accelerated” or “Accelerating”).

    oai_token-act-pair · o4-miniTriggered 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
    恆
    0.57
    ᓛ
    0.57
    fahrung
    0.56
     信息
    0.55
     information
    0.54
    횡
    0.54
     Information
    0.53
    守
    0.53
    umption
    0.53
     recipe
    0.51
    POSITIVE LOGITS
     distingu
    0.57
     পাকিস্তান
    0.56
     disting
    0.55
    UA
    0.55
    addTo
    0.54
    UEN
    0.54
     creando
    0.53
    
    0.53
    CLUSIVE
    0.53
    𝙃
    0.53
    Activations Density 0.001%

    No Known Activations