© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 12075
    Prev
    Next
    INDEX
    Explanations

    doctor-patient relationships

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

    relationships between professionals (like doctors, lawyers) and their clients/patients.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron consistently detects occurrences of the word “patient” (often in compounds or in doctor–patient 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_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
    κι
    -0.72
    Kir
    -0.69
     Amigo
    -0.69
    AndGet
    -0.66
    lossary
    -0.66
    ávat
    -0.66
     BLOCK
    -0.65
     şə
    -0.65
    Addr
    -0.65
    cciones
    -0.65
    POSITIVE LOGITS
     doctor
    2.89
     attorney
    2.41
     teacher
    2.19
    doctor
    1.98
     physician
    1.98
     client
    1.82
     employer
    1.82
    Doctor
    1.73
     Doctor
    1.71
     solicitor
    1.67
    Activations Density 0.079%

    No Known Activations