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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 29607
    Prev
    Next
    INDEX
    Explanations

    Java package names and class references, particularly related to the javax library

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

    javax programming language imports

    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
     pleaſure
    -0.74
     purpoſe
    -0.65
     Polri
    -0.60
     myſelf
    -0.58
     ſel
    -0.56
     ſtate
    -0.56
     itſelf
    -0.56
     uſe
    -0.56
    potranspiration
    -0.55
     ſever
    -0.54
    POSITIVE LOGITS
     autorytatywna
    0.60
     فريبيس
    0.57
    uxxxx
    0.52
     Bons
    0.49
    jax
    0.46
    javax
    0.46
     Jax
    0.45
     Nx
    0.45
     Dix
    0.45
    zusch
    0.44
    Activations Density 0.001%

    No Known Activations