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-2-2B
    4. Residual Stream - 65k
    5. 17-GEMMASCOPE-RES-65K
    6. 28428
    Prev
    Next
    INDEX
    Explanations

    instances of updates or revisions within content

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    UnusedPrivate
    -0.87
    SequentialGroup
    -0.68
    קישורים
    -0.63
    دانشنامهٔ
    -0.61
    FailureListener
    -0.59
    AnchorStyles
    -0.57
    VideoCapture
    -0.57
     LoggerFactory
    -0.57
    ,:,
    -0.56
    Démographie
    -0.56
    POSITIVE LOGITS
    Update
    1.34
    UPDATE
    1.33
     update
    1.26
     UPDATE
    1.23
     Update
    1.20
    update
    1.18
     updating
    0.99
     updates
    0.97
     updated
    0.96
    Updated
    0.94
    Activations Density 0.087%

    No Known Activations