© 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. 33-TRANSCODER-HP
    7. 53792
    Prev
    Next
    INDEX
    Explanations

    Technical text and code snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    -f
    -0.70
     ÑĦ
    -0.66
    _f
    -0.57
    ÑĦ
    -0.56
    ãĥķ
    -0.52
     fi
    -0.52
    -F
    -0.52
     Ùģ
    -0.51
    ãĥķãĤ£
    -0.49
    à¸Ł
    -0.47
    POSITIVE LOGITS
    xbb
    0.28
    å¾·æĭī
    0.24
    èıģ
    0.24
    èģĮä¸ļçĶŁæ¶¯
    0.23
    quo
    0.23
    ķĮ
    0.23
    noun
    0.22
    æĹĥ
    0.22
    åĬ³
    0.22
    reuse
    0.22
    Activations Density 0.104%

    No Known Activations