© 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. 31-GEMMASCOPE-2-RES-16K
    7. 5997
    Prev
    Next
    INDEX
    Explanations

    The neuron spikes on infrequent, unusual tokens—i.e. words that occur rarely in the text.

    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
    荫
    0.76
    Autres
    0.74
    ذ
    0.74
    ید
    0.72
    茱
    0.71
     జీ
    0.71
    キ
    0.70
    蝣
    0.70
     végétale
    0.70
    ী
    0.68
    POSITIVE LOGITS
     JW
    0.84
     transgender
    0.81
     grandpa
    0.79
     Maximilian
    0.79
     Mariah
    0.78
     Iraq
    0.77
    oporto
    0.76
     Congolese
    0.75
     multimillion
    0.75
     PBS
    0.74
    Activations Density 0.000%

    No Known Activations