© 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. 19842
    Prev
    Next
    INDEX
    Explanations

    glass

    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
    å·ŀå¸Ĥ
    -0.25
     Tort
    -0.25
     TORT
    -0.24
    _closure
    -0.24
    åĩºè¡Ģ
    -0.24
     Nickel
    -0.24
    è¡Ħ
    -0.24
    ä¿ĥéĶĢ
    -0.23
    juan
    -0.23
    Tony
    -0.23
    POSITIVE LOGITS
    çİ»çĴĥ
    0.87
     glass
    0.77
     Glass
    0.69
    Glass
    0.68
    glass
    0.66
    çݱ
    0.63
     glasses
    0.63
    çİ»
    0.60
    glas
    0.59
     Glas
    0.56
    Activations Density 5.468%

    No Known Activations