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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 8402
    Prev
    Next
    INDEX
    Explanations

    complex legal arguments and discussions surrounding testimony and evidence in a court context

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

    followed by "set," "apply," "mention," or "discussed"

    np_acts-logits-general · gemini-2.0-flash

    references to preceding text

    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
     zewnętrzne
    -0.55
    JNIEnv
    -0.52
     diría
    -0.51
    setupUi
    -0.48
     descrizione
    -0.47
    nerie
    -0.47
     siehe
    -0.47
     bashing
    -0.46
     sujeito
    -0.46
     mené
    -0.45
    POSITIVE LOGITS
    DeleteCommand
    0.57
     AppCompatTheme
    0.56
     ***!
    0.54
    وفاته
    0.52
     περ
    0.51
     Italijani
    0.51
     Unesco
    0.50
    ernoo
    0.50
     cog
    0.50
     signin
    0.49
    Activations Density 0.722%

    No Known Activations