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. 90079
    Prev
    Next
    INDEX
    Explanations

    objects and items associated with everyday activities

    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
    ="@+
    -0.60
     ExecuteAsync
    -0.54
     kaarangay
    -0.54
     Glut
    -0.50
    RectangleBorder
    -0.50
    SharedCtor
    -0.49
    AddTagHelper
    -0.49
    acry
    -0.47
     resourceCulture
    -0.47
    TURE
    -0.47
    POSITIVE LOGITS
    SequentialGroup
    0.41
     NavController
    0.34
    acamata
    0.32
    PullParser
    0.32
     instalado
    0.31
     aérea
    0.31
    Și
    0.31
    Şi
    0.30
     Kartoffeln
    0.30
     listed
    0.29
    Activations Density 0.039%

    No Known Activations