© 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. 12556
    Prev
    Next
    INDEX
    Explanations

    partition

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

    The neuron flags informative “content” tokens—nouns, verbs, and technical terms—rather than common function words.

    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
     पद्धतीने
    0.78
    m
    0.76
    rát
    0.71
     respons
    0.68
     পায়ের
    0.65
    ed
    0.64
     absorbance
    0.62
     পালিয়ে
    0.61
    并将
    0.61
    सिर
    0.61
    POSITIVE LOGITS
    🫢
    0.81
     cosidd
    0.80
    そもそも
    0.80
    𝐏
    0.74
    潘
    0.73
    ختلف
    0.71
    𓂃
    0.69
    そも
    0.68
    🫶
    0.68
     Polity
    0.66
    Activations Density 1.023%

    No Known Activations