© 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. 39707
    Prev
    Next
    INDEX
    Explanations

    starts of phrases/sentences

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

    The neuron never activates on any token—it isn’t detecting any pattern.

    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
     only
    -1.28
     where
    -1.23
     not
    -1.21
     if
    -1.20
     what
    -1.18
     without
    -1.16
     since
    -1.16
     over
    -1.12
     just
    -1.12
     no
    -1.10
    POSITIVE LOGITS
    
    1.04
    ativity
    1.02
     prendere
    1.01
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    1.00
    isierten
    1.00
    了一种
    0.97
    在这种
    0.96
     geleg
    0.96
    similar
    0.96
     horrid
    0.96
    Activations Density 0.000%

    No Known Activations