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. 40819
    Prev
    Next
    INDEX
    Explanations

    data structure specifications and associated metrics

    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
    ésultats
    -0.77
     autorytatywna
    -0.76
     Geiſt
    -0.76
    FormTagHelper
    -0.73
     Normdatei
    -0.72
     müſſen
    -0.71
     Numerade
    -0.71
     laſſen
    -0.69
    <unused8>
    -0.69
    <unused43>
    -0.69
    POSITIVE LOGITS
    ↵↵
    0.54
    ↵
    0.50
    0.44
     for
    0.44
     electronic
    0.42
    <
    0.41
     Egger
    0.39
    }
    0.38
     to
    0.38
     Ron
    0.37
    Activations Density 0.070%

    No Known Activations