© 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. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 9971
    Prev
    Next
    INDEX
    Explanations

    contextual words and pronouns mixed with random nouns

    oai_token-act-pair · gemini-2.0-flash

    won/wouldn't/wont

    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
    setVerticalGroup
    -0.76
    BeginInit
    -0.69
    NameInMap
    -0.61
    ContentAsync
    -0.60
    الإنجليزية
    -0.59
    WriteBarrier
    -0.53
    DockStyle
    -0.53
     [*]
    -0.51
    providedIn
    -0.50
     AssemblyProduct
    -0.49
    POSITIVE LOGITS
     won
    2.09
     Won
    1.93
    won
    1.85
    Won
    1.84
     wouldn
    1.78
     WON
    1.73
     wont
    1.70
    wouldn
    1.60
    WON
    1.46
     Wouldn
    1.45
    Activations Density 2.331%

    No Known Activations