© 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 - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 27920
    Prev
    Next
    INDEX
    Explanations

    programming-related keywords and syntax

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

    schlegel

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    DockStyle
    -0.75
    IActionResult
    -0.72
    OrBuilder
    -0.59
    DIPSETTING
    -0.59
    argout
    -0.54
     EconPapers
    -0.54
    migrationBuilder
    -0.52
    principalTable
    -0.50
     Fargo
    -0.49
    CodeDom
    -0.47
    POSITIVE LOGITS
    脚注の使い方
    0.90
    StandardCharsets
    0.59
     ErrIntOverflow
    0.59
     unmute
    0.58
     AssemblyCulture
    0.58
     vastaan
    0.54
    enservice
    0.53
     محفوظة
    0.52
     obstante
    0.52
    følgelig
    0.52
    Activations Density 0.054%

    No Known Activations