© 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. 96914
    Prev
    Next
    INDEX
    Explanations

    of

    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
    -drop
    -0.29
     nailed
    -0.28
    PLAIN
    -0.27
    æİīäºĨ
    -0.26
    lessness
    -0.26
    arme
    -0.26
    rés
    -0.26
    anced
    -0.26
    å®ĮåĸĦçļĦ
    -0.26
    less
    -0.26
    POSITIVE LOGITS
    çļĦæŶéĹ´
    0.32
     nature
    0.30
     time
    0.30
     breed
    0.28
    bre
    0.28
    åı£æ°Ķ
    0.26
    .time
    0.26
    Times
    0.25
     Time
    0.25
     Times
    0.25
    Activations Density 0.784%

    No Known Activations