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. 18-GEMMASCOPE-RES-16K
    6. 10469
    Prev
    Next
    INDEX
    Explanations

    references to personal celebrations and anniversaries

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     skall
    -0.98
     daß
    -0.78
     läßt
    -0.75
     muß
    -0.73
     müßte
    -0.66
     idéia
    -0.62
     mußte
    -0.58
     vår
    -0.56
    お勧め
    -0.54
     dignité
    -0.53
    POSITIVE LOGITS
     reportedly
    1.13
    脚注の使い方
    1.12
    ########.
    0.99
     ⁠
    0.96
     CreateTagHelper
    0.94
    abestanden
    0.93
     demnach
    0.93
     OMIT
    0.91
    tvguidetime
    0.89
     disambiguazione
    0.89
    Activations Density 0.250%

    No Known Activations