© 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. 10055
    Prev
    Next
    INDEX
    Explanations

    instances of the word "you" being able to do something

    oai_token-act-pair · gemini-2.0-flash

    possibilities

    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
     
    -0.47
     in
    -0.42
     F
    -0.41
    content
    -0.41
     perte
    -0.41
     littérature
    -0.40
     ü
    -0.40
     che
    -0.40
     er
    -0.39
     Rechte
    -0.39
    POSITIVE LOGITS
     myſelf
    0.89
     itſelf
    0.85
     $_"
    0.85
     Anſ
    0.76
    tagHelperRunner
    0.76
     ―――――
    0.73
     Muc
    0.73
    новниш
    0.73
     Efq
    0.72
     faſt
    0.72
    Activations Density 1.247%

    No Known Activations