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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 5912
    Prev
    Next
    INDEX
    Explanations

    phrases indicating motivation or encouragement to take action

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

    make a choice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     updateTime
    -0.50
    StructEnd
    -0.49
    MessageTagHelper
    -0.48
    brary
    -0.48
     currentTime
    -0.48
     HTMLElement
    -0.46
    テンツ
    -0.46
     plufieurs
    -0.46
    requently
    -0.46
     ActivatedRoute
    -0.45
    POSITIVE LOGITS
    getOption
    0.56
     Pilihan
    0.52
     go
    0.51
    ToUse
    0.46
    choix
    0.45
    take
    0.44
    odyear
    0.44
    either
    0.44
     get
    0.43
     take
    0.42
    Activations Density 0.010%

    No Known Activations