© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 60652
    Prev
    Next
    INDEX
    Explanations

    official policy and directives

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

    The neuron is tuned to detect numeric codes (digit sequences) in structured message header fields.

    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
    ĝ
    -0.75
    โครง
    -0.71
    unnitel
    -0.69
     estimation
    -0.69
    commutative
    -0.68
    daille
    -0.68
     Copyright
    -0.68
    لاس
    -0.68
     shrugged
    -0.67
     lower
    -0.67
    POSITIVE LOGITS
     policy
    1.14
     Manual
    1.10
    Manual
    1.03
     manual
    1.03
     directives
    1.00
    policy
    0.99
     manuel
    0.99
    Policy
    0.94
     Directives
    0.93
     MANUAL
    0.93
    Activations Density 0.026%

    No Known Activations