Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 68322
    Prev
    Next
    INDEX
    Explanations

    the adjective "fabulous" and its variants

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

    fab or fabulous

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
     RSI
    -0.53
    RSI
    -0.51
     Ringo
    -0.47
     Reaper
    -0.45
    getR
    -0.45
    NT
    -0.44
    qid
    -0.44
    MK
    -0.43
     Unto
    -0.43
    NC
    -0.42
    POSITIVE LOGITS
     Fabulous
    1.02
    fabulous
    0.99
     fabulous
    0.93
    Fabulous
    0.93
     Fab
    0.82
     fab
    0.79
     FAB
    0.75
     Faber
    0.72
    Fab
    0.71
    fantastic
    0.70
    Activations Density 0.004%

    No Known Activations