© 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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 93221
    Prev
    Next
    INDEX
    Explanations

    terms related to signaling pathways and gene regulation in biological processes

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

    abbreviations followed by numbers/hyphens

    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
    reactivex
    -0.39
    tamol
    -0.38
    detectChanges
    -0.36
    PerformLayout
    -0.35
     hacerte
    -0.34
    主
    -0.34
    IBOutlet
    -0.33
    ьаж
    -0.32
     CallOverrides
    -0.32
    WriteLiteral
    -0.32
    POSITIVE LOGITS
     enfans
    0.64
     gynhyrchwyd
    0.60
     ErrIntOverflow
    0.58
    XmlAccessType
    0.55
    🟤
    0.51
    ortecedor
    0.50
    delwed
    0.50
    ArrowToggle
    0.49
    addPreferredGap
    0.48
    parsedMessage
    0.48
    Activations Density 0.224%

    No Known Activations