© 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. 8204
    Prev
    Next
    INDEX
    Explanations

    measurements and values related to performance metrics

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

    documentation-style comment blocks and annotation tags in source code (API docstrings/docblocks).

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    lines in code documentation that describe function parameters (indicated by `@param`).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    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
     Westen
    -0.56
     Dooley
    -0.54
     in
    -0.52
     of
    -0.50
     Kugel
    -0.50
     Chaves
    -0.49
     wesentlichen
    -0.49
     Engle
    -0.49
     Hügel
    -0.49
     цели
    -0.49
    POSITIVE LOGITS
    PerformLayout
    0.88
    
    0.76
    tvguidetime
    0.72
    adaptiveStyles
    0.71
    UserScript
    0.71
     Infórmanos
    0.70
    parsedMessage
    0.63
    олові
    0.63
     endfor
    0.63
     
    0.62
    Activations Density 0.030%

    No Known Activations