© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 15901
    Prev
    Next
    INDEX
    Explanations

    code snippets related to generic programming and repository patterns

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

    after "<" or "'../../"

    np_acts-logits-general · gemini-2.0-flash

    models

    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
     againſt
    -0.79
    клопе
    -0.78
    DockStyle
    -0.77
     ISNI
    -0.73
     disambiguazione
    -0.72
    σιμοποι
    -0.66
     Shakspeare
    -0.66
    KURZBESCHREIBUNG
    -0.66
    醐
    -0.66
    estors
    -0.65
    POSITIVE LOGITS
     model
    1.69
     models
    1.55
    model
    1.50
    models
    1.38
     Models
    1.37
     Model
    1.37
     MODEL
    1.34
    Model
    1.33
    Models
    1.31
     modelo
    1.26
    Activations Density 0.179%

    No Known Activations