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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 22-GEMMASCOPE-RES-16K
    6. 8028
    Prev
    Next
    INDEX
    Explanations

    references to California and related geographical locations or events

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

    California locations

    np_acts-logits-general · gemini-2.0-flash

    California cities and states

    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
    RegressionTest
    -0.61
    AndEndTag
    -0.56
    ArrowToggle
    -0.49
    bonucleic
    -0.48
     Californie
    -0.48
    VYMaps
    -0.48
     Munch
    -0.48
    ‚¬
    -0.48
    tvguidetime
    -0.48
     Pokies
    -0.48
    POSITIVE LOGITS
    AddHtmlAttribute
    0.60
     Riverside
    0.53
    TagMode
    0.52
    edale
    0.50
    tractor
    0.49
     Longo
    0.48
    ViewImports
    0.48
    MockMvc
    0.47
     Fontana
    0.47
     FONT
    0.45
    Activations Density 0.152%

    No Known Activations