© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 33-TRANSCODER-HP
    7. 53046
    Prev
    Next
    INDEX
    Explanations

    Code and legal documents

    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
    Not in Any Lists

    No Comments

    Negative Logits
     togg
    -0.43
     navig
    -0.43
     ende
    -0.42
     travers
    -0.41
     Purch
    -0.41
     harb
    -0.41
     piv
    -0.41
     gras
    -0.39
     approx
    -0.39
     adher
    -0.39
    POSITIVE LOGITS
    igest
    0.28
     bleed
    0.27
    åĵī
    0.26
     escort
    0.26
    inci
    0.26
     molest
    0.25
    grow
    0.25
    abort
    0.25
    ustain
    0.25
     avail
    0.24
    Activations Density 0.198%

    No Known Activations