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

    blocks of code that handle exceptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    code segments involving exception handling and try-catch blocks.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    the beginning of 'try' blocks in code, indicating sections of code that handle exceptions.

    oai_token-act-pair · gpt-4oTriggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    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
    -------------</
    -0.63
    IsPostBack
    -0.55
     dAtA
    -0.55
    iſchen
    -0.54
     Bottom
    -0.54
     Geiſt
    -0.53
     autorytatywna
    -0.52
     externi
    -0.52
    ésultats
    -0.51
     Kommune
    -0.51
    POSITIVE LOGITS
    try
    0.81
    Try
    0.49
     try
    0.48
    TRY
    0.46
    TargetException
    0.45
     Try
    0.43
     TRY
    0.38
    pushFollow
    0.37
     trypsin
    0.35
    <thead>
    0.34
    Activations Density 0.038%

    No Known Activations