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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 5212
    Prev
    Next
    INDEX
    Explanations

    business and organizational terms related to healthcare and pharmaceutical industries

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

    Company/organization descriptions

    np_acts-logits-general · gemini-2.0-flash

    identifying what something is

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
     kasarigan
    -0.68
    Hochspringen
    -0.60
    OrWhiteSpace
    -0.60
    routeProvider
    -0.58
    principalTable
    -0.57
    ThroughAttribute
    -0.56
    ỡng
    -0.55
     LIBRARIES
    -0.53
     correspon
    -0.53
     ↓,
    -0.53
    POSITIVE LOGITS
     headquartered
    1.09
     specializes
    0.96
     founded
    0.93
     specializing
    0.92
     specialize
    0.89
     Founded
    0.86
     specialized
    0.84
    Founded
    0.83
     fundada
    0.83
     specialising
    0.82
    Activations Density 0.199%

    No Known Activations