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. 11755
    Prev
    Next
    INDEX
    Explanations

    verbs and phrases indicating possession or attributes

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

    Follows "has," "have," "had," or "got."

    np_acts-logits-general · gemini-2.0-flash

    has positive attributes

    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
    Javadoc
    -0.58
     ब्रेकडाउन
    -0.57
    sona
    -0.52
    wpdb
    -0.52
    áky
    -0.52
    argout
    -0.51
    Réponse
    -0.50
     respectively
    -0.49
    ContentLoaded
    -0.49
     ?";
    -0.49
    POSITIVE LOGITS
     really
    0.74
     lots
    0.71
     LOTS
    0.64
     REALLY
    0.62
     plenty
    0.60
     alot
    0.60
     vraiment
    0.60
     tons
    0.59
     definitely
    0.57
    really
    0.57
    Activations Density 0.161%

    No Known Activations