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 - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 698
    Prev
    Next
    INDEX
    Explanations

    references to basketball teams and players in a sports context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AndEndTag
    -0.65
     pédagogique
    -0.58
    addPreferredGap
    -0.57
    ፩
    -0.52
     students
    -0.52
    RegistryLite
    -0.50
     Studenten
    -0.50
     Students
    -0.49
     kampus
    -0.49
    WebElementEntity
    -0.48
    POSITIVE LOGITS
     NBA
    0.52
     rookie
    0.47
     franchise
    0.46
     NFL
    0.45
    franchise
    0.43
    ngdoc
    0.43
    owner
    0.41
     offseason
    0.41
     owner
    0.41
     traded
    0.41
    Activations Density 0.365%

    No Known Activations