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. 2931
    Prev
    Next
    INDEX
    Explanations

    references to scheduled tasks and intervals

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

    Regular time increments

    np_acts-logits-general · gemini-2.0-flash

    time intervals or frequency

    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
     للمعارف
    -0.65
     ويكيميديا
    -0.56
    hasErrors
    -0.52
     Máy
    -0.51
    esterday
    -0.51
     <<<<<<<<<<<<<<
    -0.51
    chyma
    -0.50
     CURIAM
    -0.49
     Vain
    -0.49
    ссу
    -0.48
    POSITIVE LOGITS
     intervals
    1.85
     interval
    1.69
     frequency
    1.56
     periodic
    1.53
    interval
    1.44
     periodicity
    1.44
     periodically
    1.41
     Intervals
    1.39
    Periodic
    1.36
     Interval
    1.36
    Activations Density 0.466%

    No Known Activations