© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 107042
    Prev
    Next
    INDEX
    Explanations

    expressive speech and language

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

    The neuron strongly activates on technical, academic discourse about speech and communication—especially terms in abstracts and papers concerning language, stuttering, and related disorders.

    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
     olhar
    -0.88
     procéder
    -0.86
     danse
    -0.85
     обло
    -0.85
    setUser
    -0.81
     mesurer
    -0.79
     NEWSLETTER
    -0.79
    掺
    -0.78
     klachten
    -0.78
     winked
    -0.77
    POSITIVE LOGITS
     speech
    3.50
    speech
    2.73
     Speech
    2.73
    Speech
    2.67
     speaking
    2.45
     speak
    2.42
     spoken
    2.34
    peech
    2.22
     communication
    2.19
     language
    2.09
    Activations Density 0.133%

    No Known Activations