© 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. 54387
    Prev
    Next
    INDEX
    Explanations

    Unintelligible text

    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
    èİ«è¿ĩäºİ
    -0.26
    male
    -0.26
    Equals
    -0.26
    jah
    -0.25
    jÄħ
    -0.24
    GreaterThan
    -0.24
    kre
    -0.24
     kre
    -0.24
     overrun
    -0.24
     Hom
    -0.24
    POSITIVE LOGITS
    æijĩ
    0.31
     Deals
    0.29
    och
    0.27
    æIJĸ
    0.27
    UR
    0.25
    æ½®
    0.24
    bands
    0.24
    templ
    0.24
    /sdk
    0.24
    sert
    0.23
    Activations Density 0.006%

    No Known Activations