© 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. 101437
    Prev
    Next
    INDEX
    Explanations

    negative conditions or restrictions related to rules and limitations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    parsedMessage
    -0.59
    zeiro
    -0.56
    WebElementEntity
    -0.56
    ineno
    -0.55
    InjectAttribute
    -0.55
    wers
    -0.54
    انجليز
    -0.53
    
    -0.53
    amship
    -0.52
    rends
    -0.52
    POSITIVE LOGITS
     Rückkehr
    0.39
     autorytatywna
    0.38
     âmes
    0.38
     mármol
    0.37
     cuarzo
    0.37
     vuelta
    0.37
    EndContext
    0.37
     člověk
    0.36
     bahía
    0.35
    initState
    0.34
    Activations Density 0.079%

    No Known Activations