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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 15913
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and sentence-ending characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
     Taktlose
    -0.79
    uxxxx
    -0.79
     Administrativna
    -0.78
     ExecuteAsync
    -0.78
    Билгалдахарш
    -0.71
     SHELL
    -0.70
     autorytatywna
    -0.70
    IntoConstraints
    -0.69
    ロウィン
    -0.68
    tagHelperRunner
    -0.68
    POSITIVE LOGITS
     duquel
    0.46
     It
    0.46
     avoient
    0.45
    The
    0.44
     If
    0.44
     nôtre
    0.42
    .
    0.42
     However
    0.41
     suivantes
    0.41
     Since
    0.40
    Activations Density 2.377%

    No Known Activations