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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 20-GEMMASCOPE-RES-16K
    6. 14629
    Prev
    Next
    INDEX
    Explanations

    themes of community, relationships, and devotion in a religious context

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

    about living a certain way

    np_acts-logits-general · gemini-2.0-flash

    daily actions and lives

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     nakalista
    -0.64
    ValueStyle
    -0.57
     surla
    -0.56
    ScopeManager
    -0.55
     contextLoads
    -0.52
     estekak
    -0.52
    GOTREF
    -0.51
     Roskov
    -0.50
     RoutedEventArgs
    -0.50
     שוליים
    -0.49
    POSITIVE LOGITS
     comportements
    0.53
     behaviors
    0.43
     ciudadanía
    0.41
     behaviours
    0.41
     Behaviors
    0.40
     actitudes
    0.39
     daily
    0.38
    ActionCreators
    0.37
    LabelTagHelper
    0.37
     consideración
    0.37
    Activations Density 0.168%

    No Known Activations