© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 22-GEMMASCOPE-RES-16K
    11. 12673
    Prev
    Next
    INDEX
    Explanations

    terms related to regulatory penalties and compliance issues

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

    fines and penalties

    np_acts-logits-general · gemini-2.0-flash

    fines penalties punishments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MigrationBuilder
    -0.80
    contentLoaded
    -0.68
    Хьажоргаш
    -0.67
    WithIOException
    -0.66
    EnableWeb
    -0.64
    esModule
    -0.64
     autorytatywna
    -0.64
     StatefulWidget
    -0.63
     voraus
    -0.61
    ValueStyle
    -0.60
    POSITIVE LOGITS
     penalty
    1.99
     penalties
    1.94
     fines
    1.92
     punishment
    1.91
     Penalty
    1.69
     Penalties
    1.67
     punishments
    1.67
     fined
    1.65
     punished
    1.64
     penal
    1.58
    Activations Density 0.791%

    No Known Activations