© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 101
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and logical operators typically seen in formulas or code

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

    Letters

    np_acts-logits-general · gemini-2.0-flash

    lists connected by 'and'

    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
     France
    -0.90
     finite
    -0.86
     Francis
    -0.86
    Facebook
    -0.85
     Facebook
    -0.84
    framework
    -0.83
    Framework
    -0.82
    Francis
    -0.82
     fibrosis
    -0.82
    Federal
    -0.82
    POSITIVE LOGITS
    AddTagHelper
    0.65
    interopRequire
    0.52
    ieteur
    0.50
    EndInit
    0.50
    modelBuilder
    0.48
    etCode
    0.47
     مشين
    0.47
    argout
    0.46
    RuleContext
    0.45
    Datuak
    0.45
    Activations Density 0.419%

    No Known Activations