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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    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. 102564
    Prev
    Next
    INDEX
    Explanations

    technical details related to programming and shell command usage

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

    Examples of Unix/Linux command-line operations using pipes, redirects, or other shell constructs that often fail when moved between different environments (interactive shell vs scripts run by cron) or when command syntax isn't properly understood for shell-specific behavior.

    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
    styleType
    -0.66
    MLLoader
    -0.56
    énario
    -0.52
     jLabel
    -0.49
     setShow
    -0.47
    represented
    -0.47
    initComponents
    -0.45
    fjspx
    -0.45
    rungsseite
    -0.44
    metries
    -0.44
    POSITIVE LOGITS
     autorytatywna
    0.54
    💈
    0.39
    Bibliograf
    0.38
     domés
    0.37
     Schluß
    0.37
     mierda
    0.36
    ItemBackground
    0.36
    enderror
    0.36
     Thanks
    0.36
     ویکی‌پدی
    0.35
    Activations Density 0.126%

    No Known Activations