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. 890
    Prev
    Next
    INDEX
    Explanations

    names of educational institutions or organizations related to special education and related topics

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

    references to specific educational centers or programs

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

    Common prefixes followed by suffixes

    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
     resourceCulture
    -0.64
     mots
    -0.51
     cardin
    -0.51
     âgées
    -0.51
    Parcelize
    -0.49
     retrouver
    -0.49
     split
    -0.49
    ίου
    -0.49
    rentiel
    -0.47
     gustar
    -0.46
    POSITIVE LOGITS
     doubtnut
    0.69
     stanovnika
    0.69
     Jdbc
    0.67
    ynchronously
    0.65
     photolibrary
    0.63
     betweenstory
    0.63
     onAnimation
    0.63
     Gif
    0.63
     strix
    0.62
    ulemon
    0.61
    Activations Density 0.105%

    No Known Activations