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-2-2B
    4. Residual Stream - 65k
    5. 17-GEMMASCOPE-RES-65K
    6. 28468
    Prev
    Next
    INDEX
    Explanations

    terms related to difficulties and challenges faced by individuals or systems

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    goku
    -0.50
    artige
    -0.49
    mög
    -0.47
    //
    -0.46
     paroisse
    -0.46
    つの
    -0.44
    gemeen
    -0.44
    IPAddress
    -0.43
    upra
    -0.42
    indeki
    -0.42
    POSITIVE LOGITS
    SourceChecksum
    0.96
    tagHelperRunner
    0.85
    ftagPool
    0.82
     الرياضيه
    0.77
    rungsseite
    0.73
    PreferredItem
    0.72
    Aiheesta
    0.72
     initComponents
    0.71
    ंदीखरीदारी
    0.68
    MessageState
    0.66
    Activations Density 0.094%

    No Known Activations