© 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. 8614
    Prev
    Next
    INDEX
    Explanations

    seeking attention or evoking emotions

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

    The neuron responds to clause‐ or list‐boundary signals—i.e. punctuation like semicolons and slashes, repeated words, and transitional adverbs that mark breaks or shifts in the text.

    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
    ور
    0.86
    Kamu
    0.84
    น
    0.79
    與
    0.79
    當中
    0.78
    ᵑ
    0.75
    λ
    0.74
     място
    0.74
    ो
    0.74
     ascertained
    0.73
    POSITIVE LOGITS
    nati
    0.89
     ইয়
    0.79
    acey
    0.79
    estrian
    0.78
     Crosby
    0.77
    sle
    0.77
    sz
    0.75
    atti
    0.73
    cex
    0.72
    sense
    0.72
    Activations Density 0.000%

    No Known Activations