© 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. Gemma-2-2B
    3. 22-GEMMASCOPE-TRANSCODER-16K
    4. 14232
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to actions of returning, going, or rehabilitating in a timeline.

    oai_token-act-pair · gemini-2.0-flash

    words starting with "re"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     again
    -2.13
    again
    -1.83
     Again
    -1.77
    Again
    -1.73
     wieder
    -1.65
     kembali
    -1.59
     novamente
    -1.53
     nuevamente
    -1.48
     AGAIN
    -1.47
     back
    -1.45
    POSITIVE LOGITS
    DoubleQuotes
    0.54
    baldi
    0.46
    
    0.44
     SRT
    0.43
    ρη
    0.41
     Berber
    0.41
     Covent
    0.41
    uitable
    0.41
     soñ
    0.41
     overrated
    0.40
    Activations Density 1.348%

    No Known Activations