© 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. 10-GEMMASCOPE-RES-131K
    7. 6951
    Prev
    Next
    INDEX
    Explanations

    I discuss

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

    This neuron detects the first word of a sentence (i.e. sentence-initial tokens, especially discourse or pronoun leads).

    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
     they
    -2.00
     они
    -1.45
    PRACTICE
    -1.42
     evitar
    -1.41
     发表于
    -1.40
     rigide
    -1.40
     afirma
    -1.37
    ﹪
    -1.37
     They
    -1.34
    🩷
    -1.33
    POSITIVE LOGITS
     what
    2.25
     those
    1.95
    1
    1.60
     but
    1.59
     that
    1.59
     not
    1.55
    3
    1.52
    4
    1.49
     this
    1.48
     our
    1.43
    Activations Density 0.062%

    No Known Activations