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. 3-GEMMASCOPE-RES-16K
    6. 10079
    Prev
    Next
    INDEX
    Explanations

    characteristics of code syntax, particularly quotations and import statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ناه
    -0.53
     estrat
    -0.50
    ciel
    -0.49
    نات
    -0.49
    cito
    -0.48
     opsi
    -0.47
     esco
    -0.47
    AMG
    -0.45
     ruban
    -0.44
    kegaard
    -0.44
    POSITIVE LOGITS
    SharedDtor
    0.88
     Audiodateien
    0.78
     /\.
    0.72
     дописавши
    0.71
    RetentionPolicy
    0.68
    ngdoc
    0.67
    phosa
    0.66
    MemoryWarning
    0.63
    存于互联网档案馆
    0.63
    expandindo
    0.63
    Activations Density 0.078%

    No Known Activations