© 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. 65808
    Prev
    Next
    INDEX
    Explanations

    code and special characters

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

    The neuron activates on all-caps identifiers or acronyms (e.g. table/column names and other uppercase 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
    .")
    
    -0.84
    。 
    -0.81
     Brooks
    -0.80
     — 
    -0.80
    =[];
    
    -0.78
    hasMany
    -0.76
     has
    -0.75
    他们
    -0.74
     had
    -0.74
    !(
    -0.74
    POSITIVE LOGITS
    ようになりました
    1.00
     queridos
    0.90
    んですよね
    0.82
    ⑅
    0.81
     vertellen
    0.81
    Rü
    0.80
    рение
    0.79
    のではなく
    0.78
     mangiare
    0.78
    Cep
    0.78
    Activations Density 0.001%

    No Known Activations