© 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. 967
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on numeric and measurement tokens (digits, units, prices, specs) in structured “spec‐sheet” or data‐list contexts.

    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
     kami
    0.62
     menambah
    0.61
     hefty
    0.60
     pundits
    0.59
     mempunyai
    0.59
     bungalows
    0.59
     _____________
    0.58
     adver
    0.58
     korzyst
    0.57
     maî
    0.57
    POSITIVE LOGITS
    Unknown
    0.89
    N
    0.80
    Undefined
    0.78
    NaN
    0.77
     Unknown
    0.77
    R
    0.77
    B
    0.75
    L
    0.75
    Unnamed
    0.74
    unknown
    0.74
    Activations Density 0.558%

    No Known Activations