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. 20-GEMMASCOPE-RES-16K
    6. 1704
    Prev
    Next
    INDEX
    Explanations

    phrases and terms related to experimental measurements and parameters in scientific contexts

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

    Mathematical or scientific formulas

    np_acts-logits-general · gemini-2.0-flash

    volume measurements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     Kla
    -0.55
     lane
    -0.53
    akir
    -0.51
     runways
    -0.51
     banner
    -0.51
     lanes
    -0.50
    レーン
    -0.49
     rainbows
    -0.48
     Spro
    -0.48
     Glen
    -0.48
    POSITIVE LOGITS
     volume
    1.51
    volume
    1.41
     Volume
    1.40
    Volume
    1.39
     VOLUME
    1.33
     volumes
    1.31
     volumen
    1.30
    volumen
    1.23
     Volumen
    1.21
     Volumes
    1.20
    Activations Density 0.701%

    No Known Activations