© 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. 32-TRANSCODER-HP
    7. 41254
    Prev
    Next
    INDEX
    Explanations

    science/technology texts

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lie
    -0.28
    Lie
    -0.27
    穹
    -0.26
    (TestCase
    -0.26
    åºŃ
    -0.26
    edin
    -0.26
    urved
    -0.25
    баÑĢ
    -0.24
    LEE
    -0.24
    masters
    -0.24
    POSITIVE LOGITS
    elas
    0.34
    åħ¶æ¬¡
    0.30
    y
    0.28
    acci
    0.27
     sob
    0.25
    amp
    0.25
    	finally
    0.24
    åŃĶ
    0.24
    yw
    0.24
     formal
    0.24
    Activations Density 0.004%

    No Known Activations