Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit 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. 20-GEMMASCOPE-RES-16K
    6. 4164
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "explain" and its variations, indicating a focus on clarification or elaboration of concepts

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

    Following "explain" (or similar words)

    np_acts-logits-general · gemini-2.0-flash

    explaining why

    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
    iibo
    -0.59
    writeField
    -0.55
    tagHelper
    -0.54
    currentColor
    -0.53
    othelioma
    -0.51
    guien
    -0.49
    forName
    -0.49
    trop
    -0.48
    findpost
    -0.47
     efectivo
    -0.47
    POSITIVE LOGITS
     why
    1.23
     explain
    0.87
     pourquoi
    0.83
    why
    0.81
     WHY
    0.78
    ation
    0.77
    lanations
    0.75
     varför
    0.75
     autorytatywna
    0.75
     reason
    0.74
    Activations Density 0.062%

    No Known Activations