© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 9-GEMMASCOPE-RES-131K
    11. 27683
    Prev
    Next
    INDEX
    Explanations

    phrases and terms related to various types of "breakers."

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot

    breaker

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     jsPsych
    -0.47
    χρι
    -0.46
    addGap
    -0.42
     Pus
    -0.41
    maxcdn
    -0.41
     Jume
    -0.40
    PreferredItem
    -0.40
    ſt
    -0.40
    ViewStyle
    -0.40
     biß
    -0.39
    POSITIVE LOGITS
     breaker
    2.42
     Breaker
    2.36
    breaker
    2.25
    Breaker
    2.17
     breakers
    2.14
    breakers
    2.02
     Breakers
    1.99
    breaking
    1.02
    Breaking
    0.91
     Breaking
    0.91
    Activations Density 0.004%

    No Known Activations