© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 34981
    Prev
    Next
    INDEX
    Explanations

    production and creation

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

    The neuron detects reporting or speech-attribution language, especially verbs and markers used in quoting someone (e.g. “said,” “tells,” direct-speech contexts).

    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
     this
    -1.45
     السياس
    -1.21
    を考える
    -1.20
     our
    -1.12
    𝐱
    -1.11
    分野
    -1.09
     zależ
    -1.09
     Benutzer
    -1.08
     mirando
    -1.05
     cucharadas
    -1.05
    POSITIVE LOGITS
     production
    1.41
     制作
    1.40
     producent
    1.31
     produs
    1.24
     प्रोड
    1.23
    producing
    1.20
     rehearsal
    1.20
     produção
    1.19
     生产
    1.16
     miele
    1.16
    Activations Density 0.035%

    No Known Activations