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. 4186
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions involving formulas and equations

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

    Fractions or ratios in mathematical expressions

    np_acts-logits-general · gemini-2.0-flash

    mathematical fractions and ratios

    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
    TestingModule
    -0.63
    PYX
    -0.61
     BrowserModule
    -0.56
    ReusableCell
    -0.54
     endblock
    -0.49
    NamedQueries
    -0.49
    transQ
    -0.48
    Bong
    -0.47
     autorytatywna
    -0.47
    المشاركات
    -0.47
    POSITIVE LOGITS
    frac
    1.46
    mfrac
    1.34
    dfrac
    1.10
     frac
    1.06
     fraction
    0.84
     fractions
    0.82
     Fraction
    0.79
     ratio
    0.74
    !*\
    0.71
    Fraction
    0.69
    Activations Density 1.080%

    No Known Activations