© 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. 64617
    Prev
    Next
    INDEX
    Explanations

    concluding words

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

    This neuron fires on words that begin a sentence or clause (i.e. the first token immediately following a period or line break).

    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
     rol
    -0.96
     zoals
    -0.95
     ciber
    -0.94
     pistes
    -0.93
    ownic
    -0.93
    ores
    -0.92
     dahil
    -0.90
     tâ
    -0.89
     fonde
    -0.89
    明显
    -0.88
    POSITIVE LOGITS
     then
    2.42
     therefore
    2.31
     thus
    1.92
     Therefore
    1.89
     so
    1.65
     thereby
    1.63
     consequently
    1.60
     entonces
    1.59
     Thus
    1.58
     поэтому
    1.57
    Activations Density 0.056%

    No Known Activations