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-IT
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 15432
    Prev
    Next
    INDEX
    Explanations

    repeated punctuation marks and possessive contractions in written dialogue

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

    Follows an apostrophe

    np_acts-logits-general · gemini-2.0-flash

    numbers and millions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     المعيارى
    -0.70
    ArgsConstructor
    -0.68
     StatefulWidget
    -0.66
    rawDesc
    -0.62
    TargetException
    -0.61
    +#+#
    -0.60
    #+#
    -0.59
    VersionUID
    -0.57
    Искәрмәләр
    -0.57
     Bewußt
    -0.56
    POSITIVE LOGITS
    RunWith
    0.35
    ConstraintMaker
    0.34
    ‌های
    0.31
     fleste
    0.31
    ujete
    0.29
    ‌هاي
    0.29
    érées
    0.28
    ọi
    0.28
    Enlaces
    0.28
    fører
    0.27
    Activations Density 1.361%

    No Known Activations