© 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. 30-TRANSCODER-HP
    7. 95015
    Prev
    Next
    INDEX
    Explanations

    migration

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    é¼ĵ
    -0.30
    eper
    -0.27
    çļĭ
    -0.26
    	day
    -0.26
    èĦĨ
    -0.25
    Day
    -0.25
     Day
    -0.25
     Rhodes
    -0.25
    lector
    -0.25
     day
    -0.25
    POSITIVE LOGITS
    è¿ģ
    0.65
     migrate
    0.59
     migration
    0.58
    migration
    0.57
     migr
    0.57
    è¿ģ移
    0.56
     migrating
    0.55
    éģ·
    0.54
     Migration
    0.52
    Migration
    0.49
    Activations Density 0.885%

    No Known Activations