© 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. 13076
    Prev
    Next
    INDEX
    Explanations

    argues assertions that

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

    The neuron fires strongly on impersonal “one”-style constructions (e.g. “One could…,” “One may…,” “One must…”) that introduce generic, formal statements.

    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
     milhares
    1.19
     cientos
    1.12
     increíble
    1.09
    𝙚
    1.02
     várias
    0.99
    𝙖
    0.99
     vários
    0.98
     flera
    0.98
     decenas
    0.98
     нескольких
    0.97
    POSITIVE LOGITS
     thereby
    0.85
     presupp
    0.73
    /
    0.70
     inasmuch
    0.70
     adduced
    0.69
     philosopher
    0.66
    '[
    0.66
     countenance
    0.64
     epistem
    0.64
    ;
    0.64
    Activations Density 0.077%

    No Known Activations