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. 10336
    Prev
    Next
    INDEX
    Explanations

    a focus on mathematical variables and symbols used in equations

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

    Mathematical or physical equations/symbols

    np_acts-logits-general · gemini-2.0-flash

    special symbols and variables

    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
     للمعارف
    -0.99
    ніципалі
    -0.80
    migrationBuilder
    -0.77
    IsPostBack
    -0.75
    OrBuilder
    -0.72
    ResumeLayout
    -0.70
    ">//
    -0.69
    SourceChecksum
    -0.68
    CppMethod
    -0.67
    GMENT
    -0.67
    POSITIVE LOGITS
    </i>
    0.51
    _{
    0.50
    </em>
    0.46
     πάντα
    0.42
    intenant
    0.42
    онер
    0.41
    </h5>
    0.40
     χρόν
    0.40
     województwie
    0.40
    agner
    0.39
    Activations Density 1.350%

    No Known Activations