Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 62867
    Prev
    Next
    INDEX
    Explanations

    patterns or themes related to social issues and perspectives on representation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    Not in Any Lists

    No Comments

    Negative Logits
     Hor
    -0.34
     HOR
    -0.33
     invid
    -0.33
    minded
    -0.32
     Horns
    -0.31
    BorderFactory
    -0.31
     minded
    -0.31
    CreateIndex
    -0.31
     Cordero
    -0.31
     decide
    -0.30
    POSITIVE LOGITS
     nahilalakip
    0.77
    oa̍t
    0.77
     Administrativna
    0.72
     autorytatywna
    0.67
     PeEnEo
    0.65
    expandindo
    0.63
    MessageOf
    0.62
    CloseOperation
    0.61
    :✨
    0.61
    principalColumn
    0.59
    Activations Density 0.060%

    No Known Activations