© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 109770
    Prev
    Next
    INDEX
    Explanations

    think or believe

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    мента
    -1.06
    стера
    -0.90
    dieren
    -0.84
    我很
    -0.79
     segera
    -0.79
    haviours
    -0.78
    lä
    -0.78
    居然
    -0.76
    Authority
    -0.76
     AUTHORITY
    -0.75
    POSITIVE LOGITS
     think
    8.00
     thinks
    5.38
    think
    5.19
    Think
    4.94
     Think
    4.91
     THINK
    3.75
    THINK
    3.41
     فکر
    3.28
    認為
    3.28
     believe
    3.20
    Activations Density 0.192%

    No Known Activations