© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 86634
    Prev
    Next
    INDEX
    Explanations

    references to changes in rules and policies affecting guidelines or regulations

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

    changes in rules and conditions

    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
    ImageContext
    -0.65
     nonUne
    -0.62
    ViewFeatures
    -0.58
    RuleContext
    -0.56
     '\\;'
    -0.53
     <>",
    -0.53
    MessageOf
    -0.52
    OGND
    -0.50
    BufferException
    -0.49
    __))
    -0.49
    POSITIVE LOGITS
     tighter
    0.55
    新たに
    0.40
     stricter
    0.40
     increased
    0.38
    制度
    0.38
    新た
    0.37
     new
    0.36
     tightened
    0.36
    新しく
    0.35
     easier
    0.35
    Activations Density 0.035%

    No Known Activations