© 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. 20-GEMMASCOPE-RES-65K
    6. 1472
    Prev
    Next
    INDEX
    Explanations

    elements related to structured data and programming syntax

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

    symbols and bracket structures related to programming, particularly in Swift

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

    scientific or programming terms

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    AntiForgeryToken
    -0.79
    HostException
    -0.67
     Theſe
    -0.57
    Sucesor
    -0.56
     HasFactory
    -0.55
     Monfieur
    -0.55
     beginnetje
    -0.55
    +#+#
    -0.55
    StructEnd
    -0.54
    yntaxException
    -0.53
    POSITIVE LOGITS
     JUGA
    0.64
     Newtonsoft
    0.54
    ConstraintMaker
    0.52
     ioutil
    0.51
    zzleHttp
    0.51
    cshtml
    0.48
     poveznice
    0.47
    ParallelGroup
    0.47
    WebControls
    0.47
    DIPSETTING
    0.45
    Activations Density 0.016%

    No Known Activations