© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 108662
    Prev
    Next
    INDEX
    Explanations

    send to an address

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

    The neuron flags tokens involved in instructions to submit or send materials (e.g. “send,” “email,” “please,” “via,” “documents”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     âges
    -0.88
     banques
    -0.86
     décennies
    -0.85
    чках
    -0.84
    Ь
    -0.82
    出了
    -0.80
     fé
    -0.80
    vább
    -0.79
    calloc
    -0.79
     familles
    -0.78
    POSITIVE LOGITS
     to
    2.50
     addressed
    1.18
     preferably
    1.05
     directed
    0.98
     dirigido
    0.96
     đến
    0.94
    preferably
    0.93
    至此
    0.92
     dirigió
    0.90
     Ді
    0.87
    Activations Density 0.027%

    No Known Activations