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

    Neuronpedia

    APICircuit 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. 10998
    Prev
    Next
    INDEX
    Explanations

    strange or odd descriptions and situations

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

    synonymous with "strange" or "odd"

    np_acts-logits-general · gemini-2.0-flash

    odd or strange things

    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
    CopyWith
    -0.49
    @[+][
    -0.47
    <>(
    -0.46
    ztály
    -0.46
    olars
    -0.45
    شرين
    -0.44
    êque
    -0.44
     داری
    -0.44
    rgba
    -0.43
    __(
    -0.43
    POSITIVE LOGITS
     étrange
    0.93
     odd
    0.88
    strange
    0.87
     Weird
    0.86
     strange
    0.86
     Strange
    0.86
    Weird
    0.85
     wierd
    0.83
     quirk
    0.83
    erweise
    0.81
    Activations Density 0.210%

    No Known Activations