© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 17934
    Prev
    Next
    INDEX
    Explanations

    Limits

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     limit
    -1.55
    éĻIJ
    -1.49
     Limit
    -1.46
    limit
    -1.45
    éĻIJåĪ¶
    -1.41
     limiting
    -1.39
    Limit
    -1.38
     limits
    -1.36
    -limit
    -1.34
     LIMIT
    -1.29
    POSITIVE LOGITS
    æį¶
    0.27
    åĨ¤
    0.27
    ihil
    0.26
     Prague
    0.25
    ä¸įå¿į
    0.25
    å«¡
    0.24
    aja
    0.24
    鸵
    0.24
    æ§Į
    0.24
    .touch
    0.24
    Activations Density 2.041%

    No Known Activations