© 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. 10-GEMMASCOPE-RES-131K
    7. 74873
    Prev
    Next
    INDEX
    Explanations

    words ending in ter

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

    The neuron strongly responds to the character sequence “ter” (i.e. tokens containing or ending in “ter”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    魴
    -2.88
    Ꭳ
    -2.64
     coloridos
    -2.45
    羮
    -2.44
    羕
    -2.39
     forn
    -2.34
    ین
    -2.34
     abou
    -2.33
    appear
    -2.28
    グルト
    -2.25
    POSITIVE LOGITS
    ter
    3.11
     What
    2.75
     When
    2.70
     There
    2.69
     otras
    2.61
     Then
    2.47
    enablog
    2.34
     That
    2.30
    )
    2.28
    しまった
    2.27
    Activations Density 0.022%

    No Known Activations