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. 10-GEMMASCOPE-RES-16K
    6. 12643
    Prev
    Next
    INDEX
    Explanations

    phrases related to programming and markup languages, particularly syntax elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    principalTable
    -0.69
    AsUp
    -0.69
     AppCompatTheme
    -0.65
    migrationBuilder
    -0.60
    ResumeLayout
    -0.53
    EditorBrowsable
    -0.49
    cal
    -0.48
    webElementXpaths
    -0.47
     Them
    -0.45
    ومت
    -0.45
    POSITIVE LOGITS
     שוליים
    0.69
     AttributeSet
    0.67
    URLException
    0.67
    loroethane
    0.67
    MLLoader
    0.64
    featureID
    0.63
     صوتيه
    0.62
    thâu
    0.61
    ftagPool
    0.60
     للمعارف
    0.60
    Activations Density 1.254%

    No Known Activations