Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 16327
    Prev
    Next
    INDEX
    Explanations

    concepts related to health, risk, and demographic characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    rawDesc
    -0.63
     Walkover
    -0.57
    Бахар
    -0.57
    SequentialGroup
    -0.57
     queſto
    -0.55
     zoude
    -0.54
    ⵌ
    -0.54
    UrlResolution
    -0.51
    ſicht
    -0.51
    הערות
    -0.50
    POSITIVE LOGITS
    üyada
    0.31
    RTCK
    0.29
     fürs
    0.28
    Autoritní
    0.27
     ekip
    0.27
    unft
    0.27
    woordig
    0.26
    ógico
    0.26
    Jereo
    0.25
    hynch
    0.25
    Activations Density 0.069%

    No Known Activations