© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 30-TRANSCODER-HP
    7. 94978
    Prev
    Next
    INDEX
    Explanations

    Risks and susceptibilities

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    -pr
    -0.93
    pr
    -0.79
     vulnerability
    -0.78
     prone
    -0.78
     vulnerable
    -0.78
     Pr
    -0.77
     susceptibility
    -0.77
    Pr
    -0.76
     Vulner
    -0.75
     susceptible
    -0.73
    POSITIVE LOGITS
    opian
    0.31
    mat
    0.27
    etric
    0.26
     mats
    0.26
    éĨ¢
    0.25
    jt
    0.25
    lasting
    0.25
    ahead
    0.24
    åĶ¬
    0.24
    认æ¸ħ
    0.24
    Activations Density 0.892%

    No Known Activations