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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 23896
    Prev
    Next
    INDEX
    Explanations

    phrases related to fishing experiences and aquatic life

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

    <start_of_turn> user

    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
    pushFollow
    -0.43
     Gegenteil
    -0.39
    样的
    -0.36
    ósitos
    -0.36
     '>=
    -0.35
    CardHeader
    -0.35
    ніципалі
    -0.33
     author
    -0.33
    poster
    -0.32
    ThroughAttribute
    -0.32
    POSITIVE LOGITS
    AddTagHelper
    0.57
     <=",
    0.52
    endphp
    0.52
     ModelExpression
    0.50
    hatic
    0.49
    Референце
    0.47
    </tfoot>
    0.46
    toBeTruthy
    0.45
     tartalomajánló
    0.44
    stylers
    0.44
    Activations Density 0.003%

    No Known Activations