© 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. 10-GEMMASCOPE-RES-131K
    4. 48435
    Prev
    Next
    INDEX
    Explanations

    reach for something

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

    The neuron activates on occurrences of “reach” (including as part of words like “reach,” “reachable,” “reaching,” etc.).

    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
     
    -2.83
     größere
    -2.41
            
    -2.39
    vantaged
    -2.34
    になりますが
    -2.33
     leichte
    -2.31
     schnelle
    -2.30
     längere
    -2.28
     Verstär
    -2.22
     große
    -2.20
    POSITIVE LOGITS
    
    2.75
    檍
    2.61
    
    2.50
    寕
    2.47
    s
    2.47
    ꠥ
    2.41
    捯
    2.38
    y
    2.36
    骘
    2.36
    菇
    2.33
    Activations Density 0.008%

    No Known Activations