© 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. 34-GEMMASCOPE-RES-131K
    7. 37777
    Prev
    Next
    INDEX
    Explanations

    searching for, accomplish this, show you how to

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

    The neuron fires strongly on ordinary English content words (nouns, verbs, adjectives, etc.) in running prose and remains silent on markup, punctuation, code, or non‐content tokens.

    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
    angering
    -1.02
    tidine
    -0.99
     .............
    -0.98
    
    -0.98
    "">
    -0.96
    ______________
    -0.96
     ™
    -0.96
     ............
    -0.94
    VersionUID
    -0.94
     食
    -0.93
    POSITIVE LOGITS
     such
    1.05
    NumOf
    1.02
    شفى
    0.97
     stuffs
    0.96
     even
    0.96
     sophisticated
    0.96
     including
    0.95
     sophistic
    0.94
     surprisingly
    0.94
     surely
    0.92
    Activations Density 0.090%

    No Known Activations