© 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 - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 102185
    Prev
    Next
    INDEX
    Explanations

    punctuation and formatting cues in the text

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

    academic/formal text starts

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    makeConstraints
    -0.53
     Italijani
    -0.48
     typelib
    -0.43
    featureID
    -0.42
    Халык
    -0.41
     Abp
    -0.41
    XmlAccessorType
    -0.41
    ThroughAttribute
    -0.40
    🟤
    -0.39
     chi̍t
    -0.39
    POSITIVE LOGITS
     [*]
    0.50
    wet
    0.44
    ViewFeatures
    0.40
    久しぶり
    0.38
     createState
    0.38
    permitAll
    0.37
     שוליים
    0.36
    CES
    0.36
    principalTable
    0.36
     CES
    0.36
    Activations Density 0.416%

    No Known Activations