© 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-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 10669
    Prev
    Next
    INDEX
    Explanations

    classes and data types commonly used in programming and software development contexts

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

    Followed by a class name or data-type

    np_acts-logits-general · gemini-2.0-flash

    dates and classes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
     isNameExpr
    -0.75
    fjspx
    -0.70
    WriteAttribute
    -0.69
    uxxxx
    -0.67
    AddTagHelper
    -0.66
    CppMethod
    -0.64
    messageInfo
    -0.63
    addCriterion
    -0.61
     invokingState
    -0.60
    ViewInit
    -0.59
    POSITIVE LOGITS
    ruptedException
    0.41
     StatelessWidget
    0.40
     cimetière
    0.38
     réalisateur
    0.35
     règne
    0.35
    
    0.35
    nologue
    0.34
    UIViewController
    0.34
     lames
    0.33
     dipendenti
    0.33
    Activations Density 0.101%

    No Known Activations