© 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. 74161
    Prev
    Next
    INDEX
    Explanations

    "to" followed by purpose verbs

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

    The neuron is primarily detecting the token "to," especially as it introduces infinitive verbs.

    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
    Elaboración
    -1.52
    Kunst
    -1.43
     Nederlands
    -1.41
    berra
    -1.40
     permett
    -1.37
     dienst
    -1.34
     péné
    -1.33
     klasse
    -1.32
     huwa
    -1.30
    -..
    -1.30
    POSITIVE LOGITS
            
    2.02
                
    1.80
    ");
    1.77
           
    1.73
              
    1.69
    ("
    1.63
                        
    1.58
    }{
    1.56
    </em>
    1.50
                    
    1.48
    Activations Density 0.051%

    No Known Activations