Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 11638
    Prev
    Next
    INDEX
    Explanations

    references and citations in scholarly texts

    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
    Personensuche
    -0.91
     يتيمه
    -0.83
     الحره
    -0.79
     autorytatywna
    -0.79
     PeEnEo
    -0.79
    postsleuth
    -0.77
    RegistryLite
    -0.76
     esternos
    -0.75
    oa̍t
    -0.73
    rungsseite
    -0.72
    POSITIVE LOGITS
    ↵↵
    0.33
     
    0.32
     stated
    0.31
    ↵
    0.29
    <eos>
    0.28
     Hence
    0.27
     rich
    0.27
     Rich
    0.26
     noted
    0.26
    .
    0.26
    Activations Density 0.475%

    No Known Activations