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. 20-GEMMASCOPE-RES-65K
    6. 1079
    Prev
    Next
    INDEX
    Explanations

    code elements related to API request handling

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

    programming-related keywords associated with API requests and fields in code

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

    field accessing with index

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     itſelf
    -0.90
     myſelf
    -0.85
     Monfieur
    -0.82
     Shakspeare
    -0.80
     Cæsar
    -0.78
     Theſe
    -0.77
     himſelf
    -0.76
     doubtnut
    -0.74
     ReactDOM
    -0.72
     Мексичка
    -0.68
    POSITIVE LOGITS
     дописавши
    0.59
    허
    0.58
    <eos>
    0.57
    RenderAtEndOf
    0.56
     للمعارف
    0.54
    richTextPanel
    0.54
     Chwiliwch
    0.54
    '])){
    
    0.54
    }}}
    0.53
    >((
    0.53
    Activations Density 0.319%

    No Known Activations