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. 1063
    Prev
    Next
    INDEX
    Explanations

    references to specific individuals, properties, or locations related to celebrity culture

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

    occurrences of function calls or method invocations in code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    tinyos
    -0.63
    twimg
    -0.59
    Geplaatst
    -0.58
    sscanf
    -0.56
    Ctor
    -0.51
    #!/
    -0.50
     '+':
    -0.49
    Obrázky
    -0.49
     Karp
    -0.48
    lud
    -0.48
    POSITIVE LOGITS
    enterOuterAlt
    0.86
    RuleContext
    0.73
    localctx
    0.71
    errHandler
    0.66
     GetEnumerator
    0.62
    onViewCreated
    0.60
    WriteTagHelper
    0.59
     ©️
    0.59
    findpost
    0.59
    CROSSTALK
    0.59
    Activations Density 0.001%

    No Known Activations