© 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. 21028
    Prev
    Next
    INDEX
    Explanations

    concepts related to understanding and describing complex situations or processes

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

    identifying or describing qualities

    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
     khớp
    -0.36
     herein
    -0.34
     individually
    -0.33
     jointly
    -0.32
     annually
    -0.30
     independently
    -0.30
    readAs
    -0.29
     separately
    -0.29
     Dumas
    -0.28
     blame
    -0.28
    POSITIVE LOGITS
     <<<<<<<<<<<<<<
    0.73
     Италијани
    0.71
    verifyException
    0.69
    
    0.68
    parsedMessage
    0.64
    
    0.63
    ########.
    0.61
    Tembelea
    0.61
     שוליים
    0.60
    archiviato
    0.60
    Activations Density 0.055%

    No Known Activations