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 - 16k
    5. 5-GEMMASCOPE-RES-16K
    6. 11674
    Prev
    Next
    INDEX
    Explanations

    the presence of specific behavioral patterns or states in written content

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     numberOfRows
    -0.54
    CIF
    -0.49
     onPage
    -0.48
     is
    -0.47
    InfoWindow
    -0.47
     Sincerely
    -0.46
     or
    -0.46
     having
    -0.45
     being
    -0.45
     keeping
    -0.45
    POSITIVE LOGITS
     $_"
    0.86
    ⓧ
    0.84
    argout
    0.80
     يتيمه
    0.79
    脚注の使い方
    0.77
    dafx
    0.77
     Efq
    0.75
     *)&
    0.75
     للمعارف
    0.74
    ^(@)
    0.73
    Activations Density 0.173%

    No Known Activations