© 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. 15769
    Prev
    Next
    INDEX
    Explanations

    The neuron is highly responsive to rare or unusually infrequent tokens (i.e. uncommon words or symbols).

    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
     would
    0.75
     MAKING
    0.71
    していました
    0.70
    ائل
    0.69
    allows
    0.69
     Watching
    0.69
     Making
    0.68
    )$.
    0.68
     Makes
    0.68
    LLY
    0.68
    POSITIVE LOGITS
    ần
    0.88
    पीय
    0.85
     hệ
    0.82
    芎
    0.82
     cần
    0.80
     evt
    0.80
     perlu
    0.80
     licha
    0.80
    荖
    0.79
    phenylsulfanyl
    0.78
    Activations Density 0.000%

    No Known Activations