© 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. 31-GEMMASCOPE-2-RES-16K
    7. 9782
    Prev
    Next
    INDEX
    Explanations

    according to, per

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

    The neuron flags passages that cite or attribute information—especially “according to” (often followed by a source or numeric reference).

    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_31_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
    遵循
    0.72
     ორი
    0.66
    ଶ
    0.61
    комен
    0.60
    สอง
    0.59
     bốn
    0.59
     رعایت
    0.58
     মেনে
    0.58
    setWalk
    0.57
    ଇ
    0.57
    POSITIVE LOGITS
     estimates
    1.01
     rumor
    0.86
     relatos
    0.86
     रिपोर्ट्स
    0.85
    estimates
    0.84
     estimations
    0.84
     cálculos
    0.81
     reports
    0.81
     据
    0.81
     insiders
    0.80
    Activations Density 0.108%

    No Known Activations