© 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. 33-TRANSCODER-HP
    7. 52512
    Prev
    Next
    INDEX
    Explanations

    Numerical or quantitative contexts

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     sixth
    -0.27
    æĮĤåı·
    -0.27
    åĩłä¹İ没æľī
    -0.27
    举
    -0.26
    `='$
    -0.26
    第åħŃ
    -0.24
     fifth
    -0.24
    suite
    -0.24
    almost
    -0.24
    åĮ
    -0.24
    POSITIVE LOGITS
    jab
    0.29
    Dig
    0.28
     dig
    0.27
    å°Ħ
    0.26
    -hooks
    0.26
     hook
    0.26
    æ¸į
    0.25
     Dig
    0.25
    èĽĶ
    0.25
    igator
    0.24
    Activations Density 0.012%

    No Known Activations