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. 18-GEMMASCOPE-RES-16K
    6. 14581
    Prev
    Next
    INDEX
    Explanations

    the concept of practice in various contexts

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

    the word "practice"

    np_acts-logits-general · gemini-2.0-flash

    practice partner, commentary

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     المعرف
    -0.67
    erequisites
    -0.65
    coration
    -0.62
    RectangleBorder
    -0.59
    urns
    -0.55
    erequisite
    -0.55
    ğraf
    -0.55
    hunters
    -0.55
     OnInit
    -0.54
    meuble
    -0.53
    POSITIVE LOGITS
     practices
    1.01
     Practices
    0.95
    Practices
    0.93
     PRACTICE
    0.93
     Practice
    0.90
     practise
    0.89
     practice
    0.85
     Practi
    0.85
    PRACTICE
    0.84
    practices
    0.84
    Activations Density 0.104%

    No Known Activations