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. 17-GEMMASCOPE-RES-65K
    6. 28426
    Prev
    Next
    INDEX
    Explanations

    numeric values and patterns related to statistics or data analysis

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    دانشنامهٔ
    -1.01
    ScopeManager
    -0.86
    
    -0.85
    تقاوى
    -0.81
    IUrlHelper
    -0.80
    oredCriteria
    -0.76
    GenerationType
    -0.75
    TypedDataSet
    -0.73
    PMailer
    -0.70
     Audiodateien
    -0.69
    POSITIVE LOGITS
     Fee
    0.46
     autorytatywna
    0.46
    omel
    0.45
     fee
    0.44
    fee
    0.44
    Fee
    0.43
    ésia
    0.42
    pojo
    0.42
     me
    0.41
     Fees
    0.41
    Activations Density 0.503%

    No Known Activations