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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 75116
    Prev
    Next
    INDEX
    Explanations

    layout attributes related to user interface design in Android

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

    horizontal alignment and orientation

    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
     Wicidata
    -0.53
    endpush
    -0.50
    basicConfig
    -0.47
    Життєпис
    -0.47
    MutableLiveData
    -0.46
    CardContent
    -0.45
    Rhestr
    -0.43
     entrado
    -0.43
    WriteBarrier
    -0.42
    脚注の使い方
    -0.42
    POSITIVE LOGITS
    SequentialGroup
    0.79
    horizontalLayout
    0.63
     horizontal
    0.57
     horizontally
    0.55
    izontal
    0.54
    HStack
    0.53
    HORIZONTAL
    0.52
    gridx
    0.50
     Horizontal
    0.49
    horiz
    0.49
    Activations Density 0.002%

    No Known Activations