Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 53295
    Prev
    Next
    INDEX
    Explanations

    references to education, recognition, and awards related to institutions or individuals

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

    Winning congratulations award

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    estors
    -0.54
     OnInit
    -0.52
    rizona
    -0.52
    colnshire
    -0.52
    HomeAsUpEnabled
    -0.51
     Provisional
    -0.49
    PrototypeOf
    -0.49
    fone
    -0.49
    orla
    -0.49
     retic
    -0.48
    POSITIVE LOGITS
    Congratulations
    1.00
     award
    0.98
     congratulations
    0.96
     Congratulations
    0.95
    受賞
    0.91
    award
    0.88
     Winning
    0.87
    Winning
    0.87
     winning
    0.86
    congratulations
    0.84
    Activations Density 0.040%

    No Known Activations