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. 31-GEMMASCOPE-RES-16K
    6. 15585
    Prev
    Next
    INDEX
    Explanations

    code related to memory management and resource disposal

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

    Methods related to object disposal

    np_acts-logits-general · gemini-2.0-flash

    object disposal and deallocation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    jmniej
    -0.35
     rest
    -0.31
     opa
    -0.29
     hat
    -0.28
    หวัด
    -0.28
     slips
    -0.28
    нологи
    -0.28
    밌
    -0.28
     missing
    -0.28
     flat
    -0.28
    POSITIVE LOGITS
    disposing
    1.52
    SharedDtor
    0.77
     ModelExpression
    0.75
     disposing
    0.73
     Dispose
    0.73
    IContainer
    0.69
     Disposal
    0.68
     dispose
    0.66
    BeginInit
    0.65
    Disposal
    0.63
    Activations Density 0.014%

    No Known Activations