© 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. 97126
    Prev
    Next
    INDEX
    Explanations

    A

    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
    acific
    -0.27
     disclaimer
    -0.26
     discount
    -0.26
    çªŁ
    -0.26
     indoor
    -0.25
    çŃĶ辩
    -0.25
     imitation
    -0.25
    èªįãĤģ
    -0.25
     Dial
    -0.25
    visitor
    -0.24
    POSITIVE LOGITS
    æ¡Ĩæŀ¶
    0.31
    (Frame
    0.26
    两
    0.26
    æľŁ
    0.25
    -Ta
    0.24
     Framework
    0.24
    åĨįåĪ°
    0.24
     FRAME
    0.24
     Fres
    0.24
    追
    0.23
    Activations Density 0.102%

    No Known Activations