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
    4. Residual Stream - 131k
    5. 32-GEMMASCOPE-RES-131K
    6. 129860
    Prev
    Next
    INDEX
    Explanations

    special characters and formats used in tagging or categorizing content

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

    for type labels

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.32.hook_resid_post
    Hook Layer
    32
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     zelfs
    -0.34
     some
    -0.34
    Gön
    -0.33
    IMPORTED
    -0.33
     with
    -0.32
     the
    -0.32
     seeing
    -0.30
     teil
    -0.29
     carefully
    -0.29
     berço
    -0.29
    POSITIVE LOGITS
     kaarangay
    0.81
    oa̍t
    0.63
    
    0.61
     Uncategorized
    0.61
    Personendaten
    0.59
    hoeddwyd
    0.57
     astronomy
    0.57
     :
    0.57
     estekak
    0.56
    лтамалар
    0.56
    Activations Density 0.035%

    No Known Activations