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. 7-GEMMASCOPE-RES-16K
    6. 14700
    Prev
    Next
    INDEX
    Explanations

    instances of the word "sometimes" and related phrases indicating frequency or occurrence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.7.hook_resid_post
    Hook Layer
    7
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     zdroj
    -0.63
    IntentFilter
    -0.61
    脚注の使い方
    -0.60
     californ
    -0.59
     위한
    -0.58
     hvid
    -0.57
    istock
    -0.57
    gewiesen
    -0.56
    ecore
    -0.56
    nuovo
    -0.56
    POSITIVE LOGITS
     sometimes
    1.86
     Sometimes
    1.80
    Sometimes
    1.79
    sometimes
    1.74
     manchmal
    1.40
     parfois
    1.39
     occasionally
    1.33
     quelquefois
    1.28
    有时
    1.28
     ibland
    1.28
    Activations Density 0.068%

    No Known Activations