© 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. 22-GEMMASCOPE-RES-131K
    7. 14835
    Prev
    Next
    INDEX
    Explanations

    potential, approximation, likelihood

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

    The neuron flags hedging or qualifying words and phrases (modal/uncertainty markers) such as “appears,” “potential,” “inspired by,” “able,” “kinda,” “probably,” and “approximately.”

    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
    oulders
    -0.93
     реально
    -0.88
     strait
    -0.86
    breathing
    -0.85
     naprawdę
    -0.85
     직
    -0.84
    preferably
    -0.84
     Crimson
    -0.84
    literally
    -0.83
    niko
    -0.82
    POSITIVE LOGITS
     but
    1.73
     because
    1.04
     or
    1.02
     what
    0.94
     then
    0.91
     good
    0.91
    ParallelGroup
    0.90
     which
    0.88
     might
    0.88
     would
    0.88
    Activations Density 0.046%

    No Known Activations