© 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. 4421
    Prev
    Next
    INDEX
    Explanations

    EH or Ech start of words

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

    This neuron responds to tokens containing the letter sequence “eh” (in any case), such as in “Eh,” “ehci,” “Ech,” or “EH.”

    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
    eseorang
    -0.74
     especialmente
    -0.67
     reliably
    -0.66
     énfasis
    -0.66
    limin
    -0.65
     Weil
    -0.65
    Ages
    -0.64
     Shannon
    -0.63
    models
    -0.63
     procedere
    -0.63
    POSITIVE LOGITS
     双
    0.80
    őzés
    0.79
    IZEN
    0.78
    FECT
    0.77
    taz
    0.76
    orted
    0.76
    hells
    0.76
    tismo
    0.75
     EH
    0.74
     ECHO
    0.74
    Activations Density 0.019%

    No Known Activations