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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1716
    Prev
    Next
    INDEX
    Explanations

    expressions of enjoyment or a positive experience

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

    Comes before an exclamation or possessive pronoun related to enjoyment

    np_acts-logits-general · gemini-2.0-flash

    enjoy followed by common words

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    ine
    -0.63
    ξύ
    -0.57
    c
    -0.50
    قدام
    -0.49
    ual
    -0.49
    netinet
    -0.48
    gens
    -0.47
    mazaki
    -0.47
     глазами
    -0.47
     Dalton
    -0.45
    POSITIVE LOGITS
     itſelf
    0.85
    ably
    0.84
     lenker
    0.82
     تضيفلها
    0.77
    ViewFeatures
    0.73
     pleinement
    0.73
     mergeFrom
    0.72
    themselves
    0.72
    AddTagHelper
    0.71
     pleasures
    0.70
    Activations Density 0.059%

    No Known Activations