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. 20-GEMMASCOPE-RES-16K
    6. 10184
    Prev
    Next
    INDEX
    Explanations

    references to vulnerable populations and issues related to socioeconomic status and ethnicity

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    tableFuture
    -0.46
     sede
    -0.44
    XmlAccessType
    -0.42
    Đ
    -0.42
     j
    -0.41
    seitigen
    -0.41
    databinding
    -0.40
    ⓧ
    -0.40
    tový
    -0.40
    DataBinding
    -0.39
    POSITIVE LOGITS
     groups
    1.87
     group
    1.70
    groups
    1.61
     Groups
    1.57
     groupes
    1.53
    group
    1.49
     GROUPS
    1.49
    Groups
    1.44
     subgroups
    1.44
     subgroup
    1.43
    Activations Density 0.680%

    No Known Activations