© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 14899
    Prev
    Next
    INDEX
    Explanations

    any color

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

    The neuron responds strongly to low‐frequency content words—i.e. rare or unusually specific tokens.

    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
     Kindle
    0.88
     PWM
    0.85
     UTM
    0.85
     SMTP
    0.83
     A
    0.82
     M
    0.82
     UT
    0.82
     Rine
    0.82
     B
    0.81
     DMP
    0.81
    POSITIVE LOGITS
    ​.
    0.87
    Goed
    0.80
    েলি
    0.78
    entities
    0.78
    Throughout
    0.76
    kosten
    0.75
    Jahr
    0.75
     molécules
    0.73
     théorie
    0.72
     করির
    0.71
    Activations Density 0.000%

    No Known Activations