© 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. 60650
    Prev
    Next
    INDEX
    Explanations

    `@echo off` or `set -e`

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

    The neuron fires on standalone imperative code tokens—shell or script commands and function‐call names (e.g. “par”, “union all”, “echo off”, “pushd”, “glLoadIdentity”).

    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
    珦
    -1.39
    蟜
    -1.37
            
    -1.34
    boxylic
    -1.32
                      
    -1.32
    を利用
    -1.27
     garis
    -1.27
    _
    -1.27
    I
    -1.26
              
    -1.23
    POSITIVE LOGITS
     пребы
    1.41
    =[]
    1.41
    。“
    1.38
     colheres
    1.36
     was
    1.36
    豋
    1.34
    헉
    1.30
     новых
    1.27
     havent
    1.27
    ⦊
    1.25
    Activations Density 0.001%

    No Known Activations