© 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. 110920
    Prev
    Next
    INDEX
    Explanations

    Harry

    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
    ä¸Ģ个éĹ®é¢ĺ
    -0.27
    çľĭå¾ħ
    -0.26
    aly
    -0.25
    ñe
    -0.25
    rador
    -0.25
    bow
    -0.24
    íĺ¼
    -0.24
    诱æĥij
    -0.24
     Emp
    -0.23
    æĭĻ
    -0.23
    POSITIVE LOGITS
    sie
    0.29
    ä¸Ģé¡¿
    0.27
    sic
    0.26
    æĸ¯é¡¿
    0.25
    (cf
    0.25
    slope
    0.25
     Levine
    0.25
    iculo
    0.24
    Appe
    0.24
    cf
    0.24
    Activations Density 0.173%

    No Known Activations