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. 61971
    Prev
    Next
    INDEX
    Explanations

    arrays being initialized or modified in code

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

    sprintf calls

    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
    "},
    
    -0.59
    }}},
    -0.59
    __':
    
    -0.57
    "):
    
    -0.55
     }}^{
    -0.54
    }{*}{
    -0.54
    بوابة
    -0.54
    '},
    
    -0.53
    "]
    
    -0.53
    ':
    
    -0.53
    POSITIVE LOGITS
    '][]
    0.57
    FlatAppearance
    0.45
    setVerticalGroup
    0.42
     typelib
    0.40
     setuptools
    0.39
    []
    0.39
    IsMutable
    0.39
     embarazada
    0.38
     retorted
    0.37
     Đó
    0.37
    Activations Density 0.003%

    No Known Activations