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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 90085
    Prev
    Next
    INDEX
    Explanations

    SQL SELECT statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
     queſta
    -0.77
    ſelben
    -0.63
    ValueStyle
    -0.60
    iſche
    -0.59
    UserScript
    -0.59
    tableFuture
    -0.57
    ſſung
    -0.57
     utafitiHapana
    -0.57
     fashiola
    -0.56
    <pad>
    -0.56
    POSITIVE LOGITS
    findpost
    0.40
     *
    0.40
    AutoField
    0.37
    readystatechange
    0.35
     initComponents
    0.33
    printStackTrace
    0.33
     published
    0.32
      
    0.32
     recently
    0.31
     totalité
    0.30
    Activations Density 0.003%

    No Known Activations