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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 91383
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and numeric data in a technical or historical context

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

    year followed by date

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    SequentialGroup
    -0.73
    таратура
    -0.70
     Photocase
    -0.69
    <unused41>
    -0.68
     queſta
    -0.68
    <unused79>
    -0.68
    [@BOS@]
    -0.68
    <unused14>
    -0.68
    <unused8>
    -0.68
    <unused3>
    -0.68
    POSITIVE LOGITS
     then
    0.51
     year
    0.49
    then
    0.47
     subsequent
    0.46
     subsequently
    0.46
     followed
    0.45
     Subsequently
    0.43
     Then
    0.43
     kemudian
    0.43
     Year
    0.40
    Activations Density 0.051%

    No Known Activations