© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 65k
    9. 20-GEMMASCOPE-RES-65K
    11. 323
    Prev
    Next
    INDEX
    Explanations

    phrases suggesting necessity or urgency

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

    expressions of emotional states and interactions

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

    conjunction or connector

    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
     EconPapers
    -1.18
    fromnode
    -1.01
     nahilalakip
    -1.01
    PerformLayout
    -0.97
    oredCriteria
    -0.94
     ſtate
    -0.93
    StoryboardSegue
    -0.91
     myſelf
    -0.91
    DockStyle
    -0.90
    featureID
    -0.90
    POSITIVE LOGITS
    and
    0.57
     and
    0.55
     or
    0.55
     racines
    0.52
     subido
    0.52
    あるいは
    0.47
     ennen
    0.47
    Архівовано
    0.46
     gomma
    0.46
     whose
    0.46
    Activations Density 0.068%

    No Known Activations