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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72923
    Prev
    Next
    INDEX
    Explanations

    clinical factors or outcomes

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

    technical or mathematical terms in scientific and programming contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates on numeric tokens and quantitative values (e.g. digits, measures, coefficients) in the text.

    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
    量
    -0.99
     “.
    -0.99
     päiv
    -0.98
    理
    -0.96
    imbawa
    -0.94
    bnf
    -0.93
    日本人
    -0.93
     I
    -0.92
    兩
    -0.91
    ſſen
    -0.90
    POSITIVE LOGITS
                          
    1.06
                   
    1.05
               
    1.02
    ==========
    1.01
             
    1.00
                 
    0.99
                         
    0.93
                       
    0.92
                     
    0.92
                           
    0.91
    Activations Density 0.039%

    No Known Activations