© 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. 24-GEMMASCOPE-2-RES-16K
    7. 15840
    Prev
    Next
    INDEX
    Explanations

    meeting specific criteria or categories

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on metadata‐style tokens that label numbered references or versions—e.g. section numbers, exception or operating‐system versions, DOIs/PMIDs, and other identifier tags accompanying numeric codes.

    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
    ২২
    0.81
    ссмотрим
    0.74
     Bure
    0.73
    i
    0.71
    }></
    0.71
    нды
    0.71
    いる
    0.69
     MND
    0.69
    í
    0.69
     Bere
    0.69
    POSITIVE LOGITS
     laiss
    0.90
    icità
    0.84
     tessuto
    0.80
    塢
    0.80
     répar
    0.78
     déclar
    0.78
     déclaration
    0.78
     fondament
    0.77
     matériaux
    0.77
     algéb
    0.76
    Activations Density 0.001%

    No Known Activations