© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 6-GEMMASCOPE-TRANSCODER-16K
    7. 13302
    Prev
    Next
    INDEX
    Explanations

    words related to emotions, mostly negative emotions

    oai_token-act-pair · gemini-2.0-flash

    scary/frightening

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_6/width_16k/average_l0_95
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.6.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
     Tears
    -0.55
     Struggle
    -0.52
     Anspruch
    -0.49
     tears
    -0.49
     syns
    -0.49
     abraz
    -0.47
    edic
    -0.47
     disability
    -0.47
     disabled
    -0.46
    DISABLED
    -0.46
    POSITIVE LOGITS
     scary
    0.80
     fright
    0.80
     scare
    0.78
     frighten
    0.78
     frightening
    0.76
     AssemblyTitle
    0.76
     LoggerFactory
    0.76
    scary
    0.74
    Scary
    0.73
     ReactDOM
    0.72
    Activations Density 1.358%

    No Known Activations