© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2264
    Prev
    Next
    INDEX
    Explanations

    The neuron fires only on the word “like,” i.e. it detects the verb “like” as used to express a preference.

    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
     번째
    0.78
     września
    0.77
     separable
    0.77
     );
    0.74
    ),(
    0.74
     hadn
    0.73
     septum
    0.72
     který
    0.72
     measurable
    0.72
     聞い
    0.72
    POSITIVE LOGITS
    k
    0.93
    ر
    0.86
    சொ
    0.80
    f
    0.78
    tob
    0.77
    什么
    0.77
    rison
    0.76
    侈
    0.75
    fass
    0.75
    栝
    0.73
    Activations Density 0.000%

    No Known Activations