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. 53246
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and limit evaluations

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

    mathematical equations and algebra simplification

    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
     =
    
    -0.60
     daß
    -0.59
    faßt
    -0.59
     "'");
    -0.58
     +
    
    -0.54
     muß
    -0.54
     läßt
    -0.54
     "")
    
    -0.54
    =
    
    -0.54
     היתה
    -0.53
    POSITIVE LOGITS
    Datuak
    1.05
     resourceCulture
    0.78
     useAppContext
    0.76
    Personensuche
    0.66
    WebElementEntity
    0.65
     ajutor
    0.65
    OCCURRED
    0.64
    enderror
    0.64
     noDo
    0.63
     pinulongan
    0.61
    Activations Density 0.026%

    No Known Activations