© 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. 2515
    Prev
    Next
    INDEX
    Explanations

    This neuron strongly activates on tokens with very high vocabulary indices (i.e. rare or late-indexed 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
    ieben
    0.84
     muut
    0.79
    ী
    0.78
    ről
    0.77
    εσ
    0.77
     tweede
    0.75
     σε
    0.75
    Ζ
    0.75
    σσ
    0.74
    za
    0.74
    POSITIVE LOGITS
     aggressively
    0.77
     robotics
    0.75
     biotechnology
    0.71
     biotech
    0.69
     HDTV
    0.68
     eliminates
    0.68
    Longrightarrow
    0.68
     didn
    0.67
    իմ
    0.66
     doesn
    0.66
    Activations Density 0.000%

    No Known Activations