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. 6893
    Prev
    Next
    INDEX
    Explanations

    phrases indicating job titles or organizational affiliations

    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
    if
    -0.51
    个
    -0.50
    ci
    -0.48
    cin
    -0.48
    ist
    -0.47
    ous
    -0.47
    se
    -0.46
    eng
    -0.46
    int
    -0.46
    ம்ப
    -0.45
    POSITIVE LOGITS
     CreateTagHelper
    1.12
     فريبيس
    0.96
    OOTDTY
    0.78
    Попис
    0.76
     estekak
    0.75
    fitrión
    0.74
     Roskov
    0.73
    ronique
    0.73
    abetes
    0.72
    yntaxException
    0.72
    Activations Density 0.539%

    No Known Activations