© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66541
    Prev
    Next
    INDEX
    Explanations

    not real or fake

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

    The neuron responds to words that signal something is false, a hoax, prank, conspiracy, or otherwise not genuine.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     Stupid
    -1.05
     praticamente
    -1.04
     ゆう
    -1.02
     プール
    -0.99
     īpa
    -0.96
     wonderfully
    -0.94
     装修
    -0.93
    uresh
    -0.92
     adorned
    -0.92
     soooo
    -0.91
    POSITIVE LOGITS
     just
    1.53
     only
    1.40
     merely
    1.38
     csak
    1.20
     hanya
    1.17
     только
    1.16
     simplemente
    1.13
     лишь
    1.09
    only
    1.05
    ϳ
    1.05
    Activations Density 0.035%

    No Known Activations