© 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. 22-TRANSCODER-HP
    7. 88377
    Prev
    Next
    INDEX
    Explanations

    Academic/Textual documents

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æīįè¡Į
    -0.32
    anan
    -0.26
     Lump
    -0.26
    obra
    -0.25
    Ùħبر
    -0.25
    á»ijt
    -0.25
    ukan
    -0.25
    structuring
    -0.24
    ÑĤен
    -0.24
    èĥ½ä¸įèĥ½
    -0.24
    POSITIVE LOGITS
     compositions
    0.26
    adh
    0.26
     entered
    0.25
    åĪĨ为
    0.25
     includes
    0.25
     consists
    0.25
    .codehaus
    0.25
    çĨµ
    0.25
    æ¯ı天éĥ½
    0.24
     divided
    0.24
    Activations Density 0.015%

    No Known Activations