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. 20-GEMMASCOPE-RES-16K
    6. 11179
    Prev
    Next
    INDEX
    Explanations

    code-related structures, such as method definitions and error handling

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

    Code snippets: Date formatting, error handling, and file operations

    np_acts-logits-general · gemini-2.0-flash

    try

    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
    +#+
    -0.81
    RetentionPolicy
    -0.80
    GEBURTSDATUM
    -0.66
     betweenstory
    -0.64
    UnusedPrivate
    -0.62
    kháu
    -0.61
    Portail
    -0.57
    FieldDescriptor
    -0.55
    WebElementEntity
    -0.55
     Référence
    -0.54
    POSITIVE LOGITS
    try
    2.70
     try
    2.26
     Try
    2.01
     tries
    1.87
    Try
    1.83
     TRY
    1.77
     tried
    1.75
    TRY
    1.71
     trying
    1.62
     attempt
    1.55
    Activations Density 0.060%

    No Known Activations