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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 10-GEMMASCOPE-RES-16K
    6. 3008
    Prev
    Next
    INDEX
    Explanations

    patterns related to regular expressions and syntax elements commonly used in coding

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     lo
    -0.49
    canActivate
    -0.48
    ({"
    -0.45
     EClass
    -0.45
     cap
    -0.44
    Портал
    -0.43
    santa
    -0.43
    )"),
    -0.42
    |/
    -0.42
     se
    -0.42
    POSITIVE LOGITS
    +#+
    0.78
    ########.
    0.78
    adaptiveStyles
    0.76
     purpoſe
    0.64
     utafitiHapana
    0.62
     Reſ
    0.62
    setupUi
    0.61
    ConstraintMaker
    0.60
     Egli
    0.58
     Majefty
    0.58
    Activations Density 0.484%

    No Known Activations