© 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. 41520
    Prev
    Next
    INDEX
    Explanations

    large numbers

    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
    æ¶Ī
    -0.27
    amerate
    -0.26
    å¿¿
    -0.25
    lement
    -0.25
    bere
    -0.25
    uner
    -0.24
    onta
    -0.24
    éĶ¥
    -0.23
     Lowe
    -0.23
     concert
    -0.23
    POSITIVE LOGITS
     thousand
    0.68
     thousands
    0.68
     hundred
    0.60
    åįĥ
    0.59
     hundreds
    0.58
    çĻ¾
    0.56
     Thousands
    0.55
     millions
    0.52
     Hundred
    0.52
     Thousand
    0.51
    Activations Density 0.006%

    No Known Activations