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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 16371
    Prev
    Next
    INDEX
    Explanations

    temporal markers and phrases indicating the passage of time

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    awtextra
    -0.56
     testSet
    -0.54
    Trichloroethane
    -0.54
    quedas
    -0.53
     Picchu
    -0.52
    irability
    -0.50
    ⤹
    -0.49
    AISSEE
    -0.48
     setName
    -0.48
     facture
    -0.47
    POSITIVE LOGITS
     snowing
    0.41
     grieved
    0.37
    あっという
    0.37
     inilah
    0.36
     enough
    0.36
    Enough
    0.35
    شهاد
    0.35
     bothering
    0.35
     time
    0.34
     everybody
    0.34
    Activations Density 0.016%

    No Known Activations