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. 20-GEMMASCOPE-RES-65K
    6. 1025
    Prev
    Next
    INDEX
    Explanations

    terms and specifications related to coding frameworks and libraries

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

    numeric values or references related to versioning and software packages

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

    .NET framework versions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     lenker
    -0.62
     صوتيه
    -0.46
     للاسماء
    -0.46
    реди
    -0.46
    ्यालय
    -0.46
    foot
    -0.46
    laceae
    -0.45
    Географи
    -0.45
     الحره
    -0.45
    SequentialGroup
    -0.44
    POSITIVE LOGITS
    DoubleQuotes
    0.69
    Gambas
    0.63
     ujednoznacz
    0.59
    WriteBarrier
    0.57
    省市镇
    0.56
    endpush
    0.52
     CreateTagHelper
    0.50
    abancı
    0.50
     Siamese
    0.49
    harusnya
    0.49
    Activations Density 0.043%

    No Known Activations