© 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. 25-TRANSCODER-HP
    7. 109319
    Prev
    Next
    INDEX
    Explanations

    code/technical documentation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    iri
    -0.31
    ["$
    -0.28
    éĶı
    -0.26
    bearing
    -0.25
     enlarg
    -0.25
    é¹Į
    -0.25
     wal
    -0.25
    appendChild
    -0.25
    çĶ³è¯ī
    -0.25
     multiple
    -0.24
    POSITIVE LOGITS
     quot
    0.29
    è§Ĩ
    0.28
     zob
    0.28
    æĽ´é«ĺ
    0.28
    eurs
    0.27
    æ®·
    0.26
    xAA
    0.26
    aktiv
    0.25
    виÑĤ
    0.24
    earn
    0.24
    Activations Density 0.011%

    No Known Activations