© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4259
    Prev
    Next
    INDEX
    Explanations

    peaks in

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

    The neuron is triggered by salient factual or technical nouns and measurements—i.e. key content words (like specific entities, symptom terms, dates, numeric values) in expository or list‐style contexts.

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

    salient domain-specific nouns and terminology, especially technical, scientific, medical, legal, or astronomical keywords.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    technical or specialized terminology in academic, medical, legal, or scientific contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
     οποίο
    0.77
    ができ
    0.77
     полноцен
    0.73
    াজি
    0.71
     become
    0.69
     regelmäßig
    0.69
     मदद
    0.67
     целый
    0.67
    诤
    0.66
    သို့
    0.65
    POSITIVE LOGITS
    ใน
    2.14
     στην
    1.90
     في
    1.86
     katika
    1.85
     nella
    1.82
     in
    1.75
     în
    1.72
     trong
    1.71
     tại
    1.69
     ở
    1.68
    Activations Density 2.244%

    No Known Activations