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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 114910
    Prev
    Next
    INDEX
    Explanations

    instances of the word "pour" and its variations

    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
     électron
    -0.55
     ainfi
    -0.54
     ennemi
    -0.53
     armé
    -0.52
     enfans
    -0.52
    กล้
    -0.50
     genoux
    -0.49
     vectorielles
    -0.49
     Siglo
    -0.49
     délai
    -0.49
    POSITIVE LOGITS
     Pour
    0.81
     pouring
    0.77
    Pour
    0.76
     poured
    0.76
    pour
    0.75
     pour
    0.73
     POUR
    0.72
     pours
    0.69
     pyro
    0.63
    pouring
    0.63
    Activations Density 0.010%

    No Known Activations