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. 122920
    Prev
    Next
    INDEX
    Explanations

    legal terminologies and references related to legislation and amendments

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

    Technical and formal text markers that denote relationships, quantities, time references, or function calls within mathematical, legal, or programming contexts.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    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
    providedIn
    -0.64
    IntoConstraints
    -0.60
     BorderSide
    -0.56
    CloseOperation
    -0.54
    TextSpan
    -0.53
     propOrder
    -0.53
    Viitteet
    -0.51
    NavController
    -0.50
     estekak
    -0.50
    ########.
    -0.49
    POSITIVE LOGITS
     мәкал
    0.36
     EXPERIMENT
    0.34
     Machado
    0.34
    蹶
    0.33
    ärä
    0.33
     experimenta
    0.32
     experimento
    0.31
    VELAND
    0.31
     Autorisations
    0.31
     experimente
    0.30
    Activations Density 0.276%

    No Known Activations