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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 8346
    Prev
    Next
    INDEX
    Explanations

    mentions of franchise-related terminology

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    franchise locations and merchandise

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    참고
    -0.52
    wet
    -0.44
    Wet
    -0.44
     Wet
    -0.43
    httphttps
    -0.42
    nope
    -0.41
    they
    -0.41
     nek
    -0.40
    Polly
    -0.40
     Ner
    -0.40
    POSITIVE LOGITS
     Franchise
    1.14
     franchise
    1.10
    Franchise
    1.04
    franchise
    0.99
     franchises
    0.98
     franquicia
    0.95
     Franch
    0.81
    franch
    0.79
     dezelve
    0.78
     zoude
    0.78
    Activations Density 0.003%

    No Known Activations