© 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. 95613
    Prev
    Next
    INDEX
    Explanations

    stress

    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
     IMPLIED
    -0.30
    Ļ¶
    -0.28
     FieldType
    -0.27
     deriv
    -0.26
     sharing
    -0.26
    REFERRED
    -0.25
    der
    -0.25
     der
    -0.25
     ph
    -0.25
    çļĦ家åºŃ
    -0.24
    POSITIVE LOGITS
    èĢIJ
    0.37
    解
    0.33
    澡
    0.29
    ucken
    0.28
    urm
    0.27
     khúc
    0.27
    Dyn
    0.27
    CHE
    0.26
    chal
    0.26
    å¿į
    0.26
    Activations Density 0.116%

    No Known Activations