© 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. 11086
    Prev
    Next
    INDEX
    Explanations

    self-reflection and assistance

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

    The neuron activates most strongly on words referring to roads, paths, or ways (literal or figurative).

    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
    t
    0.95
    न्ग
    0.85
    cü
    0.84
     IRB
    0.84
     lf
    0.82
    늡
    0.80
     rl
    0.79
    c
    0.79
     If
    0.79
     т
    0.78
    POSITIVE LOGITS
     tecnológicas
    1.00
     மற்றும்
    0.96
     ਅਤੇ
    0.95
    といった
    0.93
    そして
    0.92
    ।
    0.92
     اجمعين
    0.91
    '।
    0.91
    ประเทศไทย
    0.91
    এবং
    0.89
    Activations Density 1.728%

    No Known Activations