© 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. 22-TRANSCODER-HP
    7. 131150
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    没æľīæĥ³åĪ°
    -0.27
    texts
    -0.27
    itere
    -0.26
    PGA
    -0.26
     clad
    -0.25
    revision
    -0.25
    etti
    -0.24
    èľĪ
    -0.24
    éĿŀ常好çļĦ
    -0.24
    sov
    -0.24
    POSITIVE LOGITS
    è¿ģ移
    0.27
    	finally
    0.26
     vej
    0.25
    ä¸ĬéĻIJ
    0.25
    imen
    0.24
    ims
    0.24
     Br
    0.23
    寺
    0.23
    .allocate
    0.23
    ä¹ĭéģĵ
    0.23
    Activations Density 0.012%

    No Known Activations