© 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. 4771
    Prev
    Next
    INDEX
    Explanations

    Server communication

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

    The neuron responds to emphatic or promotional terms (e.g. “Different,” “leading,” “need”) that highlight or stress key features or benefits.

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

    low-quality or machine-generated text, particularly in promotional or instructional content.

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

    high-frequency function words that serve grammatical roles (articles, prepositions, and infinitive markers) within sentences.

    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
    يز
    0.87
     indem
    0.80
     رست
    0.76
    ूड
    0.75
    usern
    0.75
     anni
    0.75
    uatu
    0.74
    يع
    0.73
    മ
    0.73
     sien
    0.73
    POSITIVE LOGITS
     દૂર
    0.84
     Autónoma
    0.79
     watchdog
    0.77
    탁
    0.77
     المسافه
    0.75
     periphery
    0.75
     striving
    0.74
     对于
    0.73
     Atención
    0.73
     Scholar
    0.72
    Activations Density 0.000%

    No Known Activations