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 - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 7478
    Prev
    Next
    INDEX
    Explanations

    function and class definitions in code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    IonicModule
    -0.50
     Zjednoc
    -0.50
     Normdatei
    -0.50
    ...'
    -0.49
     itſelf
    -0.49
     fotográfica
    -0.49
     Shakspeare
    -0.48
     [...]
    -0.47
     Così
    -0.47
    GeneratedMessage
    -0.46
    POSITIVE LOGITS
     __
    1.06
     CreateTagHelper
    0.97
     _
    0.87
    styleType
    0.77
    MigrationBuilder
    0.75
    __":
    
    0.74
    __':
    
    0.74
    帖最后由
    0.73
    providedIn
    0.71
    ÍT
    0.69
    Activations Density 0.052%

    No Known Activations