© 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-2-9B
    7. Residual Stream - 131k
    9. 24-GEMMASCOPE-RES-131K
    11. 1411
    Prev
    Next
    INDEX
    Explanations

    references to color variations, specifically cyan and yellow

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

    colors and lines or dots

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    AddTagHelper
    -1.00
    ſcher
    -0.93
    ſchaft
    -0.89
    NameInMap
    -0.88
     nahilalakip
    -0.88
    iſchen
    -0.85
     المعيارى
    -0.85
    IUrlHelper
    -0.83
    ſſung
    -0.83
    niſſe
    -0.82
    POSITIVE LOGITS
     Socialista
    0.37
    color
    0.36
    ,
    0.34
     while
    0.34
     and
    0.34
    Color
    0.33
    ↵↵
    0.32
     whereas
    0.30
     biru
    0.29
     burbujas
    0.29
    Activations Density 0.001%

    No Known Activations