© 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. 13322
    Prev
    Next
    INDEX
    Explanations

    file writing and response

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

    The neuron fires on numeric literals in code.

    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
     kere
    -0.96
    
    -0.91
    dyž
    -0.82
    ="
    -0.81
     noch
    -0.80
     cuarentena
    -0.80
    živ
    -0.77
    viewHolder
    -0.77
     parfüm
    -0.76
     océano
    -0.75
    POSITIVE LOGITS
     Majestic
    0.88
    TAKT
    0.85
    אח
    0.85
     Græ
    0.85
     catalysis
    0.85
    ⇆
    0.84
     tranf
    0.84
    vège
    0.84
    othelioma
    0.84
     lobbies
    0.82
    Activations Density 0.005%

    No Known Activations