© 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. 4207
    Prev
    Next
    INDEX
    Explanations

    drawing conclusions

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

    The neuron detects language that expresses drawing or stating a conclusion or decision.

    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
    responseText
    -0.88
    scanf
    -0.85
    xmlns
    -0.85
    advert
    -0.84
    ška
    -0.83
    oje
    -0.81
     reprez
    -0.81
    tigkeit
    -0.81
     détaillée
    -0.80
     empieza
    -0.79
    POSITIVE LOGITS
     conclusion
    7.22
     conclude
    5.91
     conclusions
    5.63
     concluded
    5.63
    Conclusion
    5.56
     Conclusion
    5.22
    conclusion
    4.97
     concludes
    4.69
     concluding
    4.47
     Conclusions
    4.44
    Activations Density 0.113%

    No Known Activations