© 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. 11515
    Prev
    Next
    INDEX
    Explanations

    references to labels and labeling in content

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

    "label"

    np_acts-logits-general · gemini-2.0-flash

    label and labels

    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
    parsedMessage
    -0.65
    ंदीखरीदारी
    -0.59
     />";
    -0.59
    ViewInit
    -0.56
    Читайте
    -0.54
     Stande
    -0.53
     Possibility
    -0.52
    ค์
    -0.51
    ศาสตร์
    -0.50
    Portail
    -0.50
    POSITIVE LOGITS
     labels
    0.98
     Labels
    0.92
    labels
    0.90
     lbl
    0.89
     label
    0.83
    LabelTagHelper
    0.82
    Labels
    0.78
    aarrggbb
    0.77
     UILabel
    0.76
     LABEL
    0.76
    Activations Density 0.082%

    No Known Activations