© 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. 32-TRANSCODER-HP
    7. 42725
    Prev
    Next
    INDEX
    Explanations

    has

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    _linked
    -0.27
    zza
    -0.26
    åıªè¦ģæĺ¯
    -0.25
    ä»»ä½ķå½¢å¼ı
    -0.25
    èĥ¶åĽĬ
    -0.25
    è·¡
    -0.24
    CHA
    -0.24
    ä»»ä½ķ
    -0.24
     any
    -0.24
     Steph
    -0.24
    POSITIVE LOGITS
     got
    0.44
    éĩįè¦ģ
    0.38
     Got
    0.37
     important
    0.35
    Got
    0.34
     created
    0.33
    got
    0.31
     result
    0.30
    诱导
    0.30
    important
    0.29
    Activations Density 0.006%

    No Known Activations