© 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. 92872
    Prev
    Next
    INDEX
    Explanations

    seeking professional advice

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

    The neuron flags formal recommendations to involve a professional or expert (e.g. “reviewed by an expert,” “handled with the help of a trusted lawyer,” “speak with a local agent”).

    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
     allerlei
    -0.89
    Nobel
    -0.88
    lleri
    -0.88
    tacular
    -0.87
    ˏ
    -0.84
    basically
    -0.84
    ––––
    -0.82
    ihnachts
    -0.82
    mée
    -0.81
     praktische
    -0.81
    POSITIVE LOGITS
     experienced
    2.69
     professionals
    2.48
     professional
    2.45
     qualified
    2.22
     expert
    1.99
     skilled
    1.98
     experts
    1.97
     профессиона
    1.82
     trained
    1.77
    professional
    1.74
    Activations Density 0.034%

    No Known Activations