© 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. 586
    Prev
    Next
    INDEX
    Explanations

    phrases associated with claims or statements being validated or invalidated

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

    misinformation and falsehoods

    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
    AndEndTag
    -0.70
    ValueStyle
    -0.65
    ItemBackground
    -0.62
    IsMutable
    -0.61
    tagHelperRunner
    -0.60
    WebElementEntity
    -0.57
    ViewFeatures
    -0.56
    principalColumn
    -0.55
    twimg
    -0.54
    EndContext
    -0.50
    POSITIVE LOGITS
     reality
    0.44
     actually
    0.44
     reversed
    0.42
     actuality
    0.42
    reality
    0.40
     dụ
    0.38
     fic
    0.38
     realidad
    0.38
     refuted
    0.37
    Reality
    0.37
    Activations Density 0.163%

    No Known Activations