© 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. 24-GEMMASCOPE-2-RES-16K
    7. 6060
    Prev
    Next
    INDEX
    Explanations

    tsundere responses

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

    The neuron is keyed to pick out the first salient content word (i.e. the first non-function, “topic” word) that begins each new paragraph or segment.

    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_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
     armazenamento
    0.80
    ่า
    0.80
    ী
    0.80
     вот
    0.79
     części
    0.79
    니
    0.79
     خاطر
    0.77
    ين
    0.77
     информации
    0.77
     информацией
    0.77
    POSITIVE LOGITS
    s
    0.84
    yak
    0.80
    mob
    0.78
    Ми
    0.77
    НА
    0.71
    Immutable
    0.71
    acrylate
    0.71
    nud
    0.70
    nagar
    0.70
    elective
    0.69
    Activations Density 0.001%

    No Known Activations