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. 20-GEMMASCOPE-RES-131K
    6. 114588
    Prev
    Next
    INDEX
    Explanations

    declarations and manipulations of data structures, specifically buffers and arrays

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
     pleaſure
    -0.45
    saraba
    -0.45
    endphp
    -0.43
     Anſ
    -0.40
     perſon
    -0.40
     zetten
    -0.38
     Figue
    -0.38
     taj
    -0.37
    HasColumnName
    -0.36
     apellidos
    -0.36
    POSITIVE LOGITS
    yych
    0.50
    StructEnd
    0.49
     الحره
    0.48
    bufio
    0.47
     kaynağından
    0.47
    qrstuvwxyz
    0.46
     AttributeSet
    0.46
    Hentet
    0.45
    BufferSize
    0.45
    grà
    0.45
    Activations Density 0.025%

    No Known Activations