© 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. 20-GEMMASCOPE-RES-16K
    6. 15285
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and symbols related to statistical models

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

    Statistical distributions (normal, F, etc.)

    np_acts-logits-general · gemini-2.0-flash

    normal distribution

    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
     Reactivity
    -0.42
    beat
    -0.41
    atih
    -0.39
     lame
    -0.39
    ტერ
    -0.38
    YSIS
    -0.37
    woj
    -0.37
    bare
    -0.37
    Beat
    -0.37
     Salter
    -0.36
    POSITIVE LOGITS
     distribution
    1.34
     distributions
    1.21
     Distribution
    1.20
    distribution
    1.16
     DISTRIBUTION
    1.14
    Distribution
    1.11
     distributed
    1.10
     Distributions
    1.08
     distribute
    1.06
     distribución
    1.05
    Activations Density 0.945%

    No Known Activations