Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 10-GEMMASCOPE-RES-16K
    6. 13206
    Prev
    Next
    INDEX
    Explanations

    Boolean conditions and their evaluations in programming or logical statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    TagMode
    -0.87
    MemoryWarning
    -0.78
     EconPapers
    -0.71
    ंदीखरीदारी
    -0.71
    Xna
    -0.71
     @"/
    -0.70
    IUrlHelper
    -0.70
    MigrationBuilder
    -0.67
    TypedDataSet
    -0.67
    ChildScrollView
    -0.65
    POSITIVE LOGITS
     MatDialog
    0.62
    NegativeButton
    0.58
    кобрита
    0.47
     ComVisible
    0.46
    jMenuItem
    0.46
    celotti
    0.44
    airro
    0.44
    MethodManager
    0.43
    XtraBars
    0.43
    inquirer
    0.43
    Activations Density 5.420%

    No Known Activations