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-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 52985
    Prev
    Next
    INDEX
    Explanations

    key specifications or features related to products, specifically vehicles

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

    `[` `for` `if` `4`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ,:);
    -0.64
     Arhivirano
    -0.62
    ؤلاء
    -0.59
    teady
    -0.59
     végétal
    -0.59
     betweenstory
    -0.57
    principalTable
    -0.57
    }")
    
    -0.56
    ostante
    -0.56
    şiv
    -0.55
    POSITIVE LOGITS
     estekak
    0.59
     autorytatywna
    0.56
    IGraphics
    0.54
     useState
    0.52
     Escribe
    0.51
    uses
    0.51
    Enllaces
    0.49
    ритори
    0.48
     useEffect
    0.48
    useState
    0.47
    Activations Density 0.005%

    No Known Activations