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

    Neuronpedia

    APICircuit 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. 68549
    Prev
    Next
    INDEX
    Explanations

    mentions of specific names and entities related to individuals or groups

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

    names starting with Mah, Luc, Thom, Hashi, Rog

    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
     Felicia
    -0.61
    /
    -0.55
     Breen
    -0.54
    appendChild
    -0.54
    Margaret
    -0.50
    Italian
    -0.49
    Colorado
    -0.49
    Something
    -0.49
    />";
    -0.49
    %/
    -0.49
    POSITIVE LOGITS
    astify
    0.71
     nakalista
    0.69
    tagHelperRunner
    0.62
    pulseira
    0.61
    Datuak
    0.60
    TargetException
    0.59
     doccia
    0.56
    GEBURTSDATUM
    0.56
    llavero
    0.53
     chapéu
    0.52
    Activations Density 0.071%

    No Known Activations