Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 1964
    Prev
    Next
    INDEX
    Explanations

    patterns of positive integers and arithmetic expressions in code

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

    start of user turn

    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
     istrinya
    -0.50
     Armee
    -0.49
     armée
    -0.47
     suaminya
    -0.43
     importanza
    -0.43
     Administración
    -0.43
     malade
    -0.42
     linguistique
    -0.42
    yelidikan
    -0.41
     asistentes
    -0.41
    POSITIVE LOGITS
    enumi
    0.72
     Taktlose
    0.67
    FunctionFlags
    0.62
    BeginContext
    0.60
    MessageOf
    0.59
     snippetHide
    0.58
     resourceCulture
    0.57
    enumii
    0.57
     transfieras
    0.55
     estekak
    0.54
    Activations Density 0.034%

    No Known Activations