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-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 53290
    Prev
    Next
    INDEX
    Explanations

    instances of the document's start token

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

    followed by noun phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     bl
    -0.68
    الدراسه
    -0.67
     الرياضيه
    -0.66
    ğraf
    -0.64
    databind
    -0.62
     fac
    -0.61
     pageContext
    -0.61
     cac
    -0.60
     ho
    -0.60
     mang
    -0.60
    POSITIVE LOGITS
    |}{
    1.03
    }{*}{
    1.02
    <h2>
    0.95
     Escolar
    0.85
     lyre
    0.84
     pleaſure
    0.81
     ſmall
    0.81
     myſelf
    0.81
     thistle
    0.79
     AGB
    0.78
    Activations Density 0.000%

    No Known Activations