© 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. 35-TRANSCODER-HP
    7. 40300
    Prev
    Next
    INDEX
    Explanations

    study

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    客æłĪ
    -0.32
     luckily
    -0.31
     ironically
    -0.30
     squarely
    -0.30
     rightfully
    -0.30
     supposedly
    -0.30
     casually
    -0.30
     brutally
    -0.29
    !)↵↵
    -0.29
     freaking
    -0.29
    POSITIVE LOGITS
    -M
    0.33
    -P
    0.32
    -A
    0.32
    J
    0.31
    -B
    0.31
    -E
    0.31
    -R
    0.31
    -I
    0.30
    -V
    0.30
     PA
    0.30
    Activations Density 19.381%

    No Known Activations