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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 11-GEMMASCOPE-RES-16K
    6. 1813
    Prev
    Next
    INDEX
    Explanations

    references to victimization and injustice

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     myſelf
    -0.76
     Monfieur
    -0.72
     itſelf
    -0.62
     Jefus
    -0.62
     uſe
    -0.59
     klasa
    -0.57
     Reſ
    -0.57
     Chriftian
    -0.57
     uſed
    -0.57
     ſeveral
    -0.54
    POSITIVE LOGITS
     receives
    0.95
     receive
    0.88
     receiving
    0.86
     recevoir
    0.83
     received
    0.81
    received
    0.79
     Receives
    0.78
    receiving
    0.76
     Receiving
    0.74
     Receive
    0.74
    Activations Density 0.589%

    No Known Activations