© 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. 95131
    Prev
    Next
    INDEX
    Explanations

    Proper names

    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
    пÑĥ
    -0.29
    pass
    -0.28
    cale
    -0.27
    èĩ´
    -0.26
    bk
    -0.26
    ackle
    -0.26
    kea
    -0.26
    passed
    -0.26
     pass
    -0.26
    -pass
    -0.26
    POSITIVE LOGITS
    hurst
    0.27
    uracion
    0.25
    VAR
    0.25
    icio
    0.25
     indifference
    0.25
     herald
    0.24
    ederation
    0.24
    å°Ķ
    0.24
     searcher
    0.24
    urances
    0.24
    Activations Density 0.005%

    No Known Activations