© 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. 10-GEMMASCOPE-RES-131K
    7. 100415
    Prev
    Next
    INDEX
    Explanations

    purporting to claim or state

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

    The neuron responds to verbs and phrases that signal assertions or reported claims (like “professing,” “believe,” “testing,” “purporting,” “claiming,” “promote,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     pensar
    -1.09
    คิด
    -1.09
     буде
    -1.06
    dbury
    -1.05
    Why
    -1.02
     both
    -1.02
     still
    -1.02
     and
    -1.00
     +
    -1.00
     occasionally
    -0.99
    POSITIVE LOGITS
     ,'
    1.24
    却又
    1.17
     '
    1.15
     béisbol
    1.13
     same
    1.12
    eek
    1.10
    razer
    1.10
     ,"
    1.10
     Ähn
    1.09
     then
    1.05
    Activations Density 0.083%

    No Known Activations