© 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. 95000
    Prev
    Next
    INDEX
    Explanations

    code and/or foreign languages

    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
     observation
    -0.30
     Observ
    -0.29
     observing
    -0.29
     observe
    -0.29
     convers
    -0.29
     observed
    -0.28
    è§Ĥ
    -0.28
    åıįåºĶ
    -0.27
     reaction
    -0.27
    èĩªçIJĨ
    -0.27
    POSITIVE LOGITS
    ovsky
    0.26
    Maker
    0.26
     dryer
    0.26
    ano
    0.25
    åijĬè¯īä»ĸ
    0.24
    /accounts
    0.24
    ARR
    0.24
    æį©
    0.24
     Messenger
    0.24
     ferry
    0.24
    Activations Density 0.001%

    No Known Activations