© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 31-GEMMASCOPE-RES-131K
    11. 89725
    Prev
    Next
    INDEX
    Explanations

    specific phrases and sentence structures that indicate self-identification and responses to inquiries

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

    user turn starts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     possibly
    -0.40
     investor
    -0.40
     probably
    -0.38
     thus
    -0.37
     automatic
    -0.36
     Erwachsenen
    -0.36
     freshman
    -0.36
     reversed
    -0.35
    KommentareTeilen
    -0.35
    Тарихы
    -0.35
    POSITIVE LOGITS
     contextLoads
    0.61
     архивлан
    0.56
     мәкал
    0.53
     Италијани
    0.52
    tagHelperRunner
    0.52
    endregion
    0.52
     betweenstory
    0.51
    новништво
    0.50
    adaptiveStyles
    0.50
    funny
    0.50
    Activations Density 0.009%

    No Known Activations