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. 964
    Prev
    Next
    INDEX
    Explanations

    curly braces used in code

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

    instances of opening curly braces

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

    error punch

    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
    kuuta
    -0.59
     Egli
    -0.58
     transférez
    -0.54
    thâu
    -0.53
    Leben
    -0.52
     elif
    -0.51
     Treue
    -0.51
     horizons
    -0.51
     vuitton
    -0.50
    Aiheesta
    -0.50
    POSITIVE LOGITS
     error
    0.96
     Error
    0.86
     errors
    0.81
    error
    0.80
    errors
    0.78
     failure
    0.78
     ERROR
    0.76
    ERROR
    0.75
    Error
    0.75
     failed
    0.74
    Activations Density 0.023%

    No Known Activations