© 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. 31-GEMMASCOPE-RES-131K
    11. 111342
    Prev
    Next
    INDEX
    Explanations

    verbs related to actions involving consuming food and drink or communication

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

    drinking, speaking, eating, buying, answering consumed items

    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
    CppMethod
    -0.60
    windowFixed
    -0.52
    typeorm
    -0.48
     fijar
    -0.48
    ecake
    -0.48
    RenderAtEndOf
    -0.48
     ConstraintSet
    -0.47
    argli
    -0.47
     שוליים
    -0.47
    leneck
    -0.47
    POSITIVE LOGITS
    Jîn
    0.37
     them
    0.37
    Välislingid
    0.37
     externi
    0.35
    UrlResolution
    0.35
     traditional
    0.34
     gerne
    0.34
    Mo
    0.34
     Mo
    0.34
    あえて
    0.31
    Activations Density 0.097%

    No Known Activations