© 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. 20-GEMMASCOPE-RES-16K
    6. 11637
    Prev
    Next
    INDEX
    Explanations

    place names and geographical locations

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

    Geographic locations: names of nature or places

    np_acts-logits-general · gemini-2.0-flash

    natural place names

    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
    Jeografia
    -0.49
     chromosomes
    -0.48
    PrototypeOf
    -0.48
    aarrggbb
    -0.44
     dez
    -0.42
     virtuales
    -0.42
    principalColumn
    -0.41
     rozs
    -0.41
     normaux
    -0.40
     Dinosaurs
    -0.40
    POSITIVE LOGITS
     Hill
    1.00
     Creek
    0.97
     Houſe
    0.95
     Ridge
    0.93
     houſe
    0.91
     Point
    0.89
    Creek
    0.88
     Brook
    0.83
     Valley
    0.82
    Ridge
    0.82
    Activations Density 0.258%

    No Known Activations