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. 20-GEMMASCOPE-RES-131K
    6. 22463
    Prev
    Next
    INDEX
    Explanations

    categorical terms related to food items and menu categories

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    GEBURTSDATUM
    -0.44
    SequentialGroup
    -0.41
    richTextPanel
    -0.40
    forRoot
    -0.37
    principalColumn
    -0.35
    ."
    -0.35
     anz
    -0.34
     Braid
    -0.34
    ".
    -0.33
    HasMaxLength
    -0.33
    POSITIVE LOGITS
    новништво
    0.57
     Chwiliwch
    0.56
     EconPapers
    0.55
     informée
    0.54
    قایناق‌لار
    0.52
     casada
    0.51
     '\\;'
    0.50
    tagHelperRunner
    0.49
    MigrationBuilder
    0.49
     OMITBAD
    0.49
    Activations Density 0.086%

    No Known Activations