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. 70666
    Prev
    Next
    INDEX
    Explanations

    practical tips and instructional content related to using tools or platforms

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

    Instructional language offering guidance on how to use digital products, services or features, typically appearing in tutorials, documentation, or training materials.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    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
     surla
    -0.72
     Exactos
    -0.58
    thenReturn
    -0.52
    fillType
    -0.52
    dersfield
    -0.50
    complexContent
    -0.49
     disambiguazione
    -0.49
     Taktlose
    -0.48
    AddTagHelper
    -0.46
    findpost
    -0.46
    POSITIVE LOGITS
     user
    0.45
     User
    0.41
     productivity
    0.41
     suite
    0.39
     skills
    0.39
     navigation
    0.39
     Beginner
    0.37
     beginner
    0.36
    لينكات
    0.35
     voy
    0.35
    Activations Density 0.048%

    No Known Activations