© 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. 31-TRANSCODER-HP
    7. 18996
    Prev
    Next
    INDEX
    Explanations

    health care

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     pay
    -0.45
     claim
    -0.38
     claims
    -0.36
     Pay
    -0.33
    claim
    -0.33
     pays
    -0.33
    Claim
    -0.32
    igel
    -0.32
     Claim
    -0.32
    PAY
    -0.31
    POSITIVE LOGITS
    åıªè¦ģ
    0.29
     succ
    0.28
    boro
    0.28
    дан
    0.25
     TInt
    0.25
     ridden
    0.24
    è¹ĭ
    0.24
     dif
    0.24
     community
    0.24
     Lar
    0.23
    Activations Density 8.010%

    No Known Activations