© 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. 53331
    Prev
    Next
    INDEX
    Explanations

    the

    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
    age
    -0.30
    erk
    -0.27
    ages
    -0.25
    amt
    -0.25
     Pers
    -0.25
    ahkan
    -0.24
    çļĦèµĦéĩij
    -0.24
    çĪ¶åŃIJ
    -0.24
    转åħ¥
    -0.24
    amb
    -0.24
    POSITIVE LOGITS
     Gos
    0.27
    èģĨ
    0.27
    chas
    0.26
    æĪijä¸įæĥ³
    0.26
    bags
    0.25
    etten
    0.25
    (pipe
    0.25
    onian
    0.24
     Sco
    0.24
    请大家
    0.24
    Activations Density 0.011%

    No Known Activations