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-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 2054
    Prev
    Next
    INDEX
    Explanations

    references to forms and their structures in technical or scientific contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 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
     autorytatywna
    -1.02
     resourceCulture
    -0.94
    AddTagHelper
    -0.87
    RegressionTest
    -0.79
    Autoritní
    -0.79
    GEBURTSDATUM
    -0.78
    aarrggbb
    -0.75
     CreateTagHelper
    -0.74
    rungsseite
    -0.68
     Wikimedijinoj
    -0.68
    POSITIVE LOGITS
    WillAppear
    0.25
     uLocal
    0.23
    ButterKnife
    0.23
     kasarigan
    0.22
    等
    0.22
    AS
    0.22
    навли
    0.21
    <h1>
    0.21
    anglès
    0.21
    UP
    0.21
    Activations Density 35.268%

    No Known Activations