© 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. 5552
    Prev
    Next
    INDEX
    Explanations

    statistical methods and analyses related to regression and hypothesis testing

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

    Time series analysis and regression

    np_acts-logits-general · gemini-2.0-flash

    autoregressive time series analysis

    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
    onomy
    -0.61
    الحياه
    -0.55
    dealloc
    -0.55
    WriteBarrier
    -0.54
     ujednoznacz
    -0.49
     sauvages
    -0.49
    ább
    -0.49
    carian
    -0.48
    baker
    -0.47
    invokeLater
    -0.47
    POSITIVE LOGITS
     lags
    0.61
     stationary
    0.61
     series
    0.59
     seasonally
    0.59
     séries
    0.59
     lag
    0.58
     seasonal
    0.56
     residuals
    0.56
    trend
    0.56
     moving
    0.55
    Activations Density 1.164%

    No Known Activations