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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 32-GEMMASCOPE-RES-131K
    6. 129931
    Prev
    Next
    INDEX
    Explanations

    references to the concept of growth or coming of age

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

    grew up

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.32.hook_resid_post
    Hook Layer
    32
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    BagLayout
    -0.67
     مرئيه
    -0.61
    /**
    -0.55
    UseVisualStyle
    -0.52
    出版年
    -0.51
    ruik
    -0.50
    windowFixed
    -0.50
    المناصب
    -0.49
    providedIn
    -0.49
     pape
    -0.48
    POSITIVE LOGITS
     watching
    0.63
     entour
    0.55
     surrounded
    0.54
     hearing
    0.54
     hating
    0.51
     knowing
    0.47
     believing
    0.47
     listening
    0.47
     seeing
    0.44
     admiring
    0.43
    Activations Density 0.005%

    No Known Activations