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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 14257
    Prev
    Next
    INDEX
    Explanations

    introductory phrases and contextual setups in academic or informational texts

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

    user starts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     Administrativna
    -0.96
     незавершена
    -0.89
    WriteBarrier
    -0.88
    AndEndTag
    -0.87
     protoimpl
    -0.86
     Italijani
    -0.86
    تقاوى
    -0.85
     Савезне
    -0.85
     للمعارف
    -0.83
    styleType
    -0.83
    POSITIVE LOGITS
    1
    0.28
    [
    0.27
     oscu
    0.26
    alignSelf
    0.24
    ​​
    0.24
    2
    0.24
    <em>
    0.24
    0.24
     
    
    0.23
    VStack
    0.23
    Activations Density 0.039%

    No Known Activations