© 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. 53120
    Prev
    Next
    INDEX
    Explanations

    pro

    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
     Vers
    -0.32
     vers
    -0.28
    ned
    -0.28
    ape
    -0.27
     young
    -0.27
    ir
    -0.27
    åħ¼
    -0.26
    å©´åĦ¿
    -0.26
     rem
    -0.26
    ped
    -0.26
    POSITIVE LOGITS
    toi
    0.27
    ADR
    0.27
     lif
    0.26
    lide
    0.26
    volt
    0.26
    ORY
    0.26
    Life
    0.26
     bureaucr
    0.26
    åħ°èĬ±
    0.25
    éĥİ
    0.25
    Activations Density 0.016%

    No Known Activations