© 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. 53012
    Prev
    Next
    INDEX
    Explanations

    academic/therapeutic texts

    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
     cooker
    -0.28
    agrid
    -0.27
    street
    -0.27
    chia
    -0.26
    enton
    -0.25
    åŀĴ
    -0.25
    bucks
    -0.25
    redd
    -0.25
    åľī
    -0.25
    sym
    -0.25
    POSITIVE LOGITS
     analytical
    0.32
     drug
    0.32
    èį¯
    0.30
     integr
    0.30
     object
    0.28
    æ¯Ĵ
    0.27
     analytics
    0.27
     pet
    0.27
     comb
    0.27
     mod
    0.27
    Activations Density 0.029%

    No Known Activations