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. 5908
    Prev
    Next
    INDEX
    Explanations

    instances of object creation in code

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

    Instantiating Java collections

    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
    rungsseite
    -0.63
    ⋙
    -0.49
    protoimpl
    -0.47
    ModelAdmin
    -0.46
    最快更新
    -0.44
     handleMessage
    -0.43
    setupUi
    -0.43
    forChild
    -0.42
    
    -0.42
    ChildScrollView
    -0.41
    POSITIVE LOGITS
    ArrayList
    0.42
     LIRE
    0.41
    stdc
    0.41
     Ho
    0.41
    わけ
    0.40
    nskap
    0.38
     useState
    0.38
     java
    0.38
    Referências
    0.38
    telt
    0.38
    Activations Density 0.004%

    No Known Activations