© 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. 235
    Prev
    Next
    INDEX
    Explanations

    The neuron detects words beginning with the negative prefix “un-.”

    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
    ות
    0.41
     defray
    0.38
    ים
    0.37
    es
    0.37
    s
    0.36
    iye
    0.36
    konto
    0.33
    manship
    0.32
    lly
    0.32
    iya
    0.31
    POSITIVE LOGITS
     caric
    0.37
    ور
    0.36
    ethylsulfanyl
    0.34
    ött
    0.33
    소
    0.32
     ripet
    0.32
    addDigit
    0.32
    ান্স
    0.31
    デ
    0.31
    elementReference
    0.31
    Activations Density 0.000%

    No Known Activations