© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1365
    Prev
    Next
    INDEX
    Explanations

    cooking instructions and processes involving time and temperature

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

    time-related phrases and measurements (hours, minutes, days, overnight, etc.).

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    time periods or durations in cooking and laboratory procedures.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 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
    IsContent
    -0.65
     Administrativna
    -0.63
    enderror
    -0.63
    transQ
    -0.56
    contentLoaded
    -0.56
    -------
    -0.56
     Paglinawan
    -0.55
     defaultstate
    -0.54
    Personensuche
    -0.52
     zijne
    -0.52
    POSITIVE LOGITS
     overnight
    0.61
     hours
    0.50
     Overnight
    0.49
    vernight
    0.48
    Overnight
    0.46
     soak
    0.43
     heures
    0.41
     Hours
    0.40
     HOURS
    0.39
     days
    0.39
    Activations Density 0.215%

    No Known Activations