© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 40541
    Prev
    Next
    INDEX
    Explanations

    thus, leading to a consequence

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

    This neuron strongly activates on the discourse marker “Thus” at the start of sentences.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ",
    -2.58
                                
    -2.05
                   
    -2.05
                 
    -2.00
                        
    -2.00
    <
    -1.99
    が
    -1.97
    ?’
    -1.96
     calefacción
    -1.95
     появи
    -1.91
    POSITIVE LOGITS
     they
    1.89
     Those
    1.89
     Gemü
    1.87
     verhind
    1.81
     seguir
    1.80
    pecabe
    1.73
     But
    1.72
     Asked
    1.70
    cillary
    1.67
    蹋
    1.66
    Activations Density 0.006%

    No Known Activations