© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 23442
    Prev
    Next
    INDEX
    Explanations

    Peter Pan characters

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

    The neuron fires on tokens that are names or identifiers—anything like file‐path components and version numbers, acronyms, and proper nouns or title words (i.e. tokens containing uppercase letters or 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
    行星
    -0.72
     پول
    -0.71
    垛
    -0.70
     Strengthening
    -0.70
    hlt
    -0.70
     Troubleshooting
    -0.69
     柱
    -0.69
     Alab
    -0.68
     overridden
    -0.67
    Argentine
    -0.67
    POSITIVE LOGITS
     Never
    1.70
     Wendy
    1.68
     Peter
    1.63
    Peter
    1.53
    Wendy
    1.44
     Barrie
    1.34
    peter
    1.34
    PETER
    1.33
    Never
    1.32
     Hook
    1.27
    Activations Density 0.021%

    No Known Activations