© 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. 87647
    Prev
    Next
    INDEX
    Explanations

    Start of sentences

    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
    代è¨Ģ
    -0.31
    çļĦæĺ¯
    -0.29
    orado
    -0.27
    è´»
    -0.27
    åĩĢ
    -0.26
    edef
    -0.26
    ä¸įæŃ¢
    -0.26
     desn
    -0.26
    ä¸įä»ħ
    -0.25
    å½ķ
    -0.24
    POSITIVE LOGITS
     nå
    0.27
     Affero
    0.26
    快讯
    0.26
    çľ¼çķĮ
    0.25
     Getty
    0.25
    cura
    0.24
    -trained
    0.24
    annya
    0.24
    ä»İ严治åħļ
    0.24
     Hartford
    0.24
    Activations Density 0.001%

    No Known Activations