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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8208
    Prev
    Next
    INDEX
    Explanations

    locations and geographic features

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

    Location after preposition/determiner

    np_acts-logits-general · gemini-2.0-flash

    city names and political entities

    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
     Evansville
    -0.53
     Mino
    -0.51
     Calabria
    -0.50
     Cadiz
    -0.48
     Cramer
    -0.48
    Dund
    -0.48
     Galveston
    -0.47
     Durban
    -0.47
     Crosse
    -0.47
    Niagara
    -0.46
    POSITIVE LOGITS
     Delhi
    1.09
     Capital
    0.94
     capital
    0.92
     capitale
    0.92
    Delhi
    0.92
     Beijing
    0.92
     CAPITAL
    0.91
    Capital
    0.87
    capital
    0.86
     दिल्ली
    0.85
    Activations Density 0.157%

    No Known Activations