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. 7357
    Prev
    Next
    INDEX
    Explanations

    coding-related actions and functions, particularly related to saving and handling data

    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
    PreferredItem
    -0.44
    isser
    -0.44
    RTGC
    -0.42
     esternos
    -0.40
     manche
    -0.36
     Pole
    -0.36
     abbre
    -0.35
     pole
    -0.34
     mens
    -0.34
    就这么
    -0.33
    POSITIVE LOGITS
    MigrationBuilder
    0.76
     kasarigan
    0.63
     beginnetje
    0.55
    itinéraire
    0.54
    ConstraintMaker
    0.54
    queryInterface
    0.54
     Taktlose
    0.52
    oredCriteria
    0.48
     useAppContext
    0.48
    makeConstraints
    0.47
    Activations Density 0.005%

    No Known Activations