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. 53087
    Prev
    Next
    INDEX
    Explanations

    elements related to programming and AJAX calls

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

    code elements and structure

    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
     défaut
    -0.52
     ‎
    -0.50
     Foxx
    -0.49
    &&\
    -0.45
    doc
    -0.45
    AspNetCore
    -0.45
    ́n
    -0.45
    harp
    -0.45
    AutoField
    -0.44
    ]={
    -0.44
    POSITIVE LOGITS
    anInt
    1.11
    
    0.76
    GEBURTSDATUM
    0.73
    LookAnd
    0.72
    aClass
    0.72
    Obfuscated
    0.66
    AndEndTag
    0.66
    disposing
    0.63
    XmlType
    0.61
    tonode
    0.59
    Activations Density 0.004%

    No Known Activations