© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 64599
    Prev
    Next
    INDEX
    Explanations

    introducing specific details or explanations

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

    This neuron detects sentence beginnings, strongly activating on the first word of each sentence.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     also
    -2.06
     також
    -1.71
     также
    -1.65
     then
    -1.62
     também
    -1.51
     these
    -1.46
     and
    -1.41
     également
    -1.40
     también
    -1.36
    这样
    -1.29
    POSITIVE LOGITS
     bowiem
    1.47
     totiž
    1.17
     infatti
    1.17
     частности
    1.15
     nämlich
    1.13
     Pertama
    1.10
     zwar
    1.09
     namelijk
    1.07
    он
    1.06
     Rather
    1.05
    Activations Density 0.062%

    No Known Activations