Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 21-GEMMASCOPE-RES-16K
    6. 2079
    Prev
    Next
    INDEX
    Explanations

    code syntax elements or structures within programming contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AndEndTag
    -0.96
     المعيارى
    -0.93
     nakalista
    -0.85
     typelib
    -0.81
    PreferredItem
    -0.80
    WriteTagHelper
    -0.76
    出版年
    -0.75
     disambiguazione
    -0.74
     lenker
    -0.71
    Chham
    -0.66
    POSITIVE LOGITS
     CommonModule
    0.47
     privata
    0.47
     contemporanea
    0.46
    loat
    0.46
     crisi
    0.45
    CardModule
    0.45
    nesi
    0.44
     dovuto
    0.43
     tranquillo
    0.42
    debounce
    0.41
    Activations Density 0.186%

    No Known Activations