© 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. 33-TRANSCODER-HP
    7. 54717
    Prev
    Next
    INDEX
    Explanations

    scientific literature search

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    .less
    -0.27
    pite
    -0.25
    -answer
    -0.25
    anship
    -0.25
    å¤©åĽ½
    -0.24
    稱
    -0.24
    _fu
    -0.24
    åĵŀ
    -0.24
     Claim
    -0.24
    Claim
    -0.23
    POSITIVE LOGITS
    _PE
    0.29
    åĮĹæŀģ
    0.27
    avana
    0.27
    Roll
    0.26
    ç»ĵåIJĪ
    0.25
    åĵģä½į
    0.25
     Pur
    0.25
    ä¸Ģæľ¬
    0.25
    aves
    0.24
    Ro
    0.24
    Activations Density 0.080%

    No Known Activations