© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 65k
    9. 20-GEMMASCOPE-RES-65K
    11. 3958
    Prev
    Next
    INDEX
    Explanations

    references to a particular code structure or syntax, specifically related to variables and associations within a programming context

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot

    SB, SB Architects, SBIN_PATH, sb->, sb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    ThroughAttribute
    -0.74
    InitVars
    -0.66
    oredCriteria
    -0.62
     الاطلاع
    -0.61
     Winaray
    -0.61
    jsii
    -0.60
    nocache
    -0.59
     استنادى
    -0.58
    TagHelpers
    -0.58
    Jeografia
    -0.58
    POSITIVE LOGITS
     sb
    0.91
     Sb
    0.90
     SB
    0.89
    Sb
    0.84
    SB
    0.75
    sb
    0.72
     SBI
    0.63
     SCN
    0.59
    EndOf
    0.57
    cyd
    0.57
    Activations Density 0.017%

    No Known Activations