© 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. 18113
    Prev
    Next
    INDEX
    Explanations

    quickly

    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
    Not in Any Lists

    No Comments

    Negative Logits
     immediately
    -0.48
    ç«ĭåį³
    -0.39
     Immediately
    -0.37
     immediate
    -0.36
    Immediately
    -0.36
     hadn
    -0.33
     immedi
    -0.32
     isn
    -0.32
    Immediate
    -0.32
     hasn
    -0.31
    POSITIVE LOGITS
     moves
    0.31
     move
    0.28
    ç¦ģ
    0.28
    éĻįåĪ°
    0.28
    moves
    0.28
     moving
    0.27
    s
    0.27
    moving
    0.27
     mover
    0.27
    meld
    0.26
    Activations Density 0.846%

    No Known Activations