© 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. 96799
    Prev
    Next
    INDEX
    Explanations

    Dates and times

    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
     loop
    -0.28
    IPA
    -0.27
    //}}
    -0.25
    å°Ķ
    -0.25
     minim
    -0.25
    循
    -0.25
    -packages
    -0.24
     centered
    -0.24
     wager
    -0.24
    rego
    -0.24
    POSITIVE LOGITS
     late
    0.35
    early
    0.35
     early
    0.35
    æĹ©æľŁ
    0.34
    late
    0.31
    æĻļæľŁ
    0.28
    Early
    0.28
    letic
    0.27
     Early
    0.27
    .isPresent
    0.27
    Activations Density 0.114%

    No Known Activations