© 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. 3845
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on phrases calling out “lying to the FCC,” i.e. questions accusing someone of misleading the Federal Communications Commission.

    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
    𝐫
    0.83
    ރ
    0.79
    lexer
    0.78
    aft
    0.73
    вшего
    0.72
    вших
    0.71
     മുൻ
    0.71
    ▉
    0.70
    faith
    0.70
    ipfs
    0.70
    POSITIVE LOGITS
     procurar
    0.81
     yea
    0.76
     Lembrando
    0.73
     Humanity
    0.73
     Remember
    0.71
     ministry
    0.71
     ต้อง
    0.71
     hopeful
    0.70
     crusade
    0.70
     McE
    0.70
    Activations Density 0.000%

    No Known Activations