© 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. 43605
    Prev
    Next
    INDEX
    Explanations

    Google

    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
    äºĴ
    -0.27
    ird
    -0.26
    ('{}
    -0.26
    åĪĹ
    -0.26
    emed
    -0.26
    alted
    -0.26
    su
    -0.25
    å¸Ī
    -0.25
     tweeted
    -0.25
    師
    -0.25
    POSITIVE LOGITS
    rouw
    0.28
    antz
    0.27
    èĤĭ
    0.26
     rational
    0.25
     Serge
    0.25
    èħ¹
    0.23
    ç¾İ好
    0.23
     formation
    0.23
    èĭ±æł¼
    0.23
    ças
    0.23
    Activations Density 0.008%

    No Known Activations