© 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. 24-TRANSCODER-HP
    7. 149235
    Prev
    Next
    INDEX
    Explanations

    experiments and techniques

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     recursion
    -0.31
    强大çļĦ
    -0.27
     randomness
    -0.26
    otechnology
    -0.25
     pumps
    -0.25
    ÑĤÑıж
    -0.25
     techniques
    -0.25
     convolution
    -0.25
    äºij计ç®Ĺ
    -0.25
     Unified
    -0.24
    POSITIVE LOGITS
    æĺ¯ä¸Ģ家
    0.34
    ä¸»å¼ł
    0.27
    èĢ»
    0.25
    åŁºæľ¬æĥħåĨµ
    0.25
    awn
    0.25
    @section
    0.25
    ä¸ĩåħĥ以ä¸Ĭ
    0.24
     advocate
    0.24
    .Pages
    0.24
    ICI
    0.24
    Activations Density 0.910%

    No Known Activations