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-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1032
    Prev
    Next
    INDEX
    Explanations

    mathematical notation and symbols

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

    mathematical notation and symbols related to sequences and limits

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

    mathematical notation and variables

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    tvguidetime
    -0.79
    ConstraintMaker
    -0.77
     تضيفلها
    -0.77
    AppMethodBeat
    -0.73
    XmlAccessorType
    -0.71
     estimés
    -0.71
     tartalomajánló
    -0.71
    </thead>
    -0.70
     للاسماء
    -0.69
    RegressionTest
    -0.69
    POSITIVE LOGITS
     MessageBoxIcon
    0.72
    XmlAccessType
    0.52
    liferay
    0.46
    </strong>
    0.41
     IMDG
    0.40
    UIFont
    0.40
    Ғ
    0.39
    IMPORTED
    0.39
    [...]
    0.39
    glyphicon
    0.39
    Activations Density 1.194%

    No Known Activations