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. 18-GEMMASCOPE-RES-16K
    6. 10060
    Prev
    Next
    INDEX
    Explanations

    references to significant figures or events in Black history and prominent personalities related to Black issues or culture

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     pinulongan
    -1.02
     незавершена
    -1.01
    IntoConstraints
    -0.91
    ValueStyle
    -0.77
    +#+#
    -0.76
    脚注の使い方
    -0.74
    tagHelperRunner
    -0.73
    TestingModule
    -0.71
     HasFactory
    -0.71
    󠁢
    -0.71
    POSITIVE LOGITS
    spreis
    0.52
     تكبرها
    0.49
     the
    0.49
    XMLSchema
    0.48
    üdis
    0.44
    Twas
    0.44
     accounts
    0.43
     cuentas
    0.43
     B
    0.43
    忘了
    0.42
    Activations Density 1.996%

    No Known Activations