© 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. 119561
    Prev
    Next
    INDEX
    Explanations

    articles and conjunctions in text, indicating a focus on structural language elements

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

    the March Empire Mixing

    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
     الحره
    -0.46
    setVerticalGroup
    -0.46
    ValueGeneration
    -0.44
    TagMode
    -0.40
    reportWebVitals
    -0.40
    ViewFeatures
    -0.39
    sizeCache
    -0.39
     femei
    -0.39
     ब्रेकडाउन
    -0.39
    原始内容存档
    -0.37
    POSITIVE LOGITS
    PositiveButton
    0.47
    Contribu
    0.43
    Still
    0.42
    astify
    0.42
     convain
    0.42
     Still
    0.41
     tiens
    0.40
    Chronic
    0.40
     inqui
    0.39
     nakalista
    0.39
    Activations Density 0.059%

    No Known Activations