© 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. 20-TRANSCODER-HP
    7. 130471
    Prev
    Next
    INDEX
    Explanations

    Relative clauses

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    itch
    -0.31
    uster
    -0.28
    çĪ±
    -0.27
    ide
    -0.27
    è·Łéļı
    -0.26
    fluid
    -0.26
    ÑĢÑı
    -0.26
    ç§įæ¤į
    -0.25
    åĽŀå¤į
    -0.25
    tr
    -0.25
    POSITIVE LOGITS
     octave
    0.29
    eba
    0.29
    onth
    0.27
    ogenerated
    0.27
    .removeFrom
    0.26
    elman
    0.26
    /oct
    0.25
    ãģĪãģ°
    0.25
    lick
    0.25
     omission
    0.25
    Activations Density 1.773%

    No Known Activations