© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 18-GEMMASCOPE-RES-16K
    11. 11345
    Prev
    Next
    INDEX
    Explanations

    patterns related to data formatting and structure, particularly in scientific or technical contexts

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

    "3" after various tokens

    np_acts-logits-general · gemini-2.0-flash

    three

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ⓧ
    -0.65
     useStyles
    -0.61
    SBATCH
    -0.58
    carus
    -0.54
    GTCX
    -0.52
     onCancelled
    -0.52
     Tenth
    -0.51
    oneofs
    -0.50
    interopRequire
    -0.49
    jRadioButton
    -0.49
    POSITIVE LOGITS
    3
    1.46
     Third
    1.11
     Three
    1.10
     three
    1.09
    ３
    1.05
     ketiga
    1.04
    three
    1.02
     THREE
    1.01
     third
    1.00
     trois
    0.99
    Activations Density 1.220%

    No Known Activations