© 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. 30-TRANSCODER-HP
    7. 96396
    Prev
    Next
    INDEX
    Explanations

    mutations and disease

    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
    Not in Any Lists

    No Comments

    Negative Logits
     offline
    -0.27
    éħ¬
    -0.26
     itself
    -0.26
    essional
    -0.25
     Preview
    -0.24
    Targets
    -0.24
    offline
    -0.24
    ISP
    -0.24
     positivity
    -0.24
    _targets
    -0.24
    POSITIVE LOGITS
    fault
    0.34
     faulty
    0.32
     flaw
    0.31
    .changed
    0.30
    opleft
    0.29
     fault
    0.28
     changed
    0.28
     flawed
    0.28
    åħĦ
    0.27
    пÑĥÑĤ
    0.27
    Activations Density 0.884%

    No Known Activations