© 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. 11704
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on the very first content word after a block or paragraph break.

    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
    mem
    0.76
    methods
    0.70
    sphere
    0.69
    techn
    0.68
    いわ
    0.68
    s
    0.68
     verwenden
    0.67
    yu
    0.66
    criteria
    0.65
    electric
    0.64
    POSITIVE LOGITS
     dwind
    0.86
     ନ
    0.78
    ਾਰੇ
    0.78
     buổi
    0.77
     erupted
    0.75
     strewn
    0.75
    বক
    0.75
     vacated
    0.74
     définitive
    0.74
     tumult
    0.73
    Activations Density 0.000%

    No Known Activations