© 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. 19716
    Prev
    Next
    INDEX
    Explanations

    access

    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
     access
    -0.54
    access
    -0.53
     Access
    -0.52
    -access
    -0.52
    Access
    -0.50
     accesses
    -0.50
     acces
    -0.49
    .access
    -0.48
    ãĤ¢ãĤ¯ãĤ»ãĤ¹
    -0.46
    (access
    -0.44
    POSITIVE LOGITS
     not
    0.41
    ible
    0.31
    flix
    0.31
    _not
    0.31
    ibs
    0.29
    ow
    0.29
    not
    0.29
    	not
    0.28
     Not
    0.28
    ib
    0.28
    Activations Density 0.257%

    No Known Activations