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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 15
    Prev
    Next
    INDEX
    Explanations

    references to significant political and legal events or figures

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

    references to economic, corporate, and business contexts like companies, stores, and financial activities.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    phrases relating to specific events, dates, or measurements.

    oai_token-act-pair · gpt-4oTriggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    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
    createServer
    -0.53
     للاسماء
    -0.48
    DIPSETTING
    -0.48
     برانيه
    -0.47
    XtraEditors
    -0.47
    AnimationsModule
    -0.46
    ">—
    -0.45
    enciaga
    -0.45
    hsan
    -0.44
     cherchés
    -0.43
    POSITIVE LOGITS
     worldwide
    0.48
     nationwide
    0.48
    hyrchwyd
    0.47
    лтемелер
    0.44
     elsewhere
    0.43
     North
    0.42
     locally
    0.42
     regionally
    0.42
     nationally
    0.41
     near
    0.40
    Activations Density 0.587%

    No Known Activations