© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 19872
    Prev
    Next
    INDEX
    Explanations

    code, foreign text

    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
    Negative Logits
    standard
    -0.28
    åĪĽ
    -0.27
    aurant
    -0.26
     Pad
    -0.26
    äºĨä¸Ģä¸ĭ
    -0.26
     xmlDoc
    -0.25
    çĢ¹
    -0.25
    Pad
    -0.25
     Mont
    -0.25
    utes
    -0.25
    POSITIVE LOGITS
    cej
    0.26
    èĦĨ
    0.25
    esson
    0.24
    æľĢ好ä¸įè¦ģ
    0.24
    åį¡è½¦
    0.23
    ContentType
    0.23
     quickest
    0.23
    Accessory
    0.23
    illac
    0.23
    çķª
    0.23
    Activations Density 0.112%

    No Known Activations