© 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. 59475
    Prev
    Next
    INDEX
    Explanations

    words starting with cog or cogn

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

    The neuron activates on the character sequence “cog,” i.e. tokens containing the substring “cog” (as in cognition, cogent, cognate, cogman, etc.).

    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
    gettime
    -0.78
     taste
    -0.74
     nun
    -0.72
     langage
    -0.71
     بیان
    -0.71
     Pir
    -0.71
     nada
    -0.70
     Sagrada
    -0.70
     ara
    -0.69
    回合
    -0.69
    POSITIVE LOGITS
     cog
    1.49
    Cog
    1.40
     Cog
    1.34
     cogniz
    1.05
    cog
    1.01
    cogn
    0.94
     cogn
    0.92
     kog
    0.83
     Cognitive
    0.82
    kog
    0.82
    Activations Density 0.015%

    No Known Activations