© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 15006
    Prev
    Next
    INDEX
    Explanations

    This neuron detects high-level discourse markers—common function/content words (e.g. after, what, some, they, mean, would, requires, I use) that introduce or organize clauses.

    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
     morrer
    0.83
     binoculars
    0.81
     selfies
    0.78
     majest
    0.76
    𝚑
    0.74
     milhares
    0.73
     visitas
    0.72
     groceries
    0.71
     voordat
    0.71
     Britney
    0.70
    POSITIVE LOGITS
    思路
    0.81
    类似
    0.72
    所谓的
    0.69
    P
    0.69
    所谓
    0.66
     주로
    0.64
    近年来
    0.64
     envisaged
    0.64
    C
    0.64
    Tính
    0.63
    Activations Density 0.285%

    No Known Activations