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. 1031
    Prev
    Next
    INDEX
    Explanations

    document sections that start with a special token indicating the beginning of a new section or paragraph

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

    elements related to charming and musical characters in a narrative context

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

    positive adjectives

    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
     poveznice
    -0.56
    vejte
    -0.54
    setVerticalGroup
    -0.52
     coscienza
    -0.51
     cesse
    -0.51
     nacido
    -0.48
     pouvoirs
    -0.48
     solidarité
    -0.46
     MenuView
    -0.45
     spedizione
    -0.44
    POSITIVE LOGITS
    charming
    0.72
    delight
    0.71
     delightful
    0.69
     charming
    0.69
     handsome
    0.69
     elegant
    0.68
    ]='\
    0.68
    tonsoft
    0.68
     goodly
    0.66
    setHorizontal
    0.66
    Activations Density 0.035%

    No Known Activations