© 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. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 18999
    Prev
    Next
    INDEX
    Explanations

    news/academic writing

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     stated
    -0.57
     indicated
    -0.53
     mentioned
    -0.52
     mention
    -0.52
    æıIJåĪ°
    -0.50
    表示
    -0.49
     mentioning
    -0.48
     Mention
    -0.48
     expressed
    -0.46
     indicate
    -0.46
    POSITIVE LOGITS
     alarm
    0.35
    rightness
    0.29
     trace
    0.26
    èĨĢ
    0.26
     alarms
    0.26
     Trace
    0.24
    arg
    0.24
     Tell
    0.24
     skins
    0.24
     Sen
    0.24
    Activations Density 5.389%

    No Known Activations