© 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. 127932
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on tokens that introduce main-topic or intent statements—i.e. the opening words of summary or “purpose” sentences (like “The…,” “I would like to…,” “what…hopes to…,” “the primary use case,” etc.).

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    describing challenges and advice

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    Negative Logits
     потім
    -0.91
     більш
    -0.88
     Jasa
    -0.87
    北欧
    -0.87
     Housekeeping
    -0.81
    ਾਰ
    -0.81
     przecież
    -0.81
     exuberant
    -0.81
     Wydawnictwo
    -0.81
    Более
    -0.80
    POSITIVE LOGITS
     challenges
    1.34
    Challenges
    1.13
     biggest
    1.09
    Describe
    1.06
     Challenges
    1.05
     standout
    1.05
     advice
    1.00
     destaque
    1.00
    Advice
    0.98
     destacar
    0.97
    Activations Density 0.025%

    No Known Activations