© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 5123
    Prev
    Next
    INDEX
    Explanations

    terms related to construction and building activities

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

    "Construction" followed by a category or field

    np_acts-logits-general · gemini-2.0-flash

    construction and industry

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     <<<<<<<<<<<<<<
    -0.66
     ReactDOM
    -0.63
    RTHOOK
    -0.60
     abbiano
    -0.60
    thenReturn
    -0.60
    tvguidetime
    -0.59
    וויק
    -0.58
    ГЛА
    -0.58
    CompilerServices
    -0.57
    antMatchers
    -0.56
    POSITIVE LOGITS
    arbeiter
    0.60
    urally
    0.59
    BUILD
    0.58
    builds
    0.56
     Arbeiter
    0.56
    build
    0.55
     Vikipedi
    0.54
    onics
    0.53
     build
    0.52
     BUILD
    0.51
    Activations Density 0.107%

    No Known Activations