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-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1192
    Prev
    Next
    INDEX
    Explanations

    terms related to proactive measures and discussions

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

    details related to physical confrontation and self-defense situations

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

    proactive and reactive actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    GeoNames
    -0.59
     boson
    -0.52
    consulté
    -0.52
     jsPsych
    -0.49
     urbanization
    -0.46
     preghi
    -0.46
    Normdaten
    -0.45
     langage
    -0.44
    IonicModule
    -0.44
     Peu
    -0.43
    POSITIVE LOGITS
     reactive
    1.55
     proactive
    1.53
     proactively
    1.29
     Reactive
    1.24
    Reactive
    1.24
    reactive
    1.23
     defensive
    1.21
     preventive
    1.17
     preventative
    1.15
     Preventive
    1.14
    Activations Density 0.103%

    No Known Activations