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. 92418
    Prev
    Next
    INDEX
    Explanations

    phrases related to exclusive offers and rewards

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     vítima
    -0.46
     hoeft
    -0.45
     pernas
    -0.44
    TEntity
    -0.43
    ScopeManager
    -0.43
     يتيمه
    -0.43
     víctima
    -0.42
     lavoratori
    -0.42
     víctimas
    -0.42
     piernas
    -0.42
    POSITIVE LOGITS
     bonuses
    0.51
     bonus
    0.50
     incentives
    0.49
     rewards
    0.49
     surprises
    0.49
    bonus
    0.46
     Incentives
    0.45
    incenti
    0.44
    Награды
    0.43
    rewards
    0.41
    Activations Density 0.008%

    No Known Activations