© 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. 34-GEMMASCOPE-RES-131K
    7. 116378
    Prev
    Next
    INDEX
    Explanations

    drug approval and licensing

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

    FDA-approved medications and pharmaceutical compounds.

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

    The neuron is a “frequency detector,” activating strongly on very common tokens in the body text—especially high‐frequency function words (e.g. is, been, by, in, FDA, approved) and numeric digits.

    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
    Proposals
    -0.84
     Ub
    -0.84
     şöyle
    -0.83
    gidos
    -0.81
    ows
    -0.79
     SELECTION
    -0.78
    Jumbo
    -0.78
     אביב
    -0.78
    setupUi
    -0.77
     Carrefour
    -0.77
    POSITIVE LOGITS
     approval
    3.17
     approved
    3.06
     approve
    2.36
     approvals
    2.28
    approved
    2.27
     FDA
    2.25
     Approved
    2.20
     Approval
    2.19
    Approved
    2.11
    approval
    2.08
    Activations Density 0.212%

    No Known Activations