Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 15512
    Prev
    Next
    INDEX
    Explanations

    technical terms and processes related to manufacturing and material processing

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

    Machinery, automation, and industrial processes

    np_acts-logits-general · gemini-2.0-flash

    automatic processing machines

    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
     hilt
    -0.67
     réve
    -0.61
     Boating
    -0.58
     lenker
    -0.56
    embroidered
    -0.56
    ifest
    -0.56
     compass
    -0.55
     democrática
    -0.55
     nahilalakip
    -0.55
     ligiloj
    -0.54
    POSITIVE LOGITS
     conveyor
    0.88
     machines
    0.73
     machine
    0.72
     Machines
    0.63
     Machine
    0.62
     machinery
    0.60
     automatic
    0.58
     tph
    0.57
     automated
    0.57
     throughput
    0.56
    Activations Density 0.624%

    No Known Activations