© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 235
    Prev
    Next
    INDEX
    Explanations

    This neuron lights up on relatively rare, content‐bearing words—especially proper nouns and technical or topic‐specific terms—rather than common function words.

    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
     Shit
    0.42
    しくは
    0.41
    ロップ
    0.40
    ギュア
    0.39
     Yogurt
    0.39
     ArrayRef
    0.38
     FileName
    0.38
    ジ
    0.38
     Verificar
    0.37
    ット
    0.37
    POSITIVE LOGITS
    —
    0.61
     –
    0.58
     —
    0.58
    .
    0.53
    ,”
    0.51
    .¹
    0.50
    ​—
    0.49
     --
    0.49
    ¹.
    0.49
    ,
    0.48
    Activations Density 9.603%

    No Known Activations