© 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. 107085
    Prev
    Next
    INDEX
    Explanations

    different topics

    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
    ание
    -0.34
     Datum
    -0.30
    éĺµ
    -0.30
     ваÑĢианÑĤ
    -0.27
     statistic
    -0.26
    alink
    -0.26
    Datum
    -0.26
    ä¸Ģå¹ħ
    -0.26
    (tm
    -0.25
    ение
    -0.25
    POSITIVE LOGITS
     phases
    0.36
     thresholds
    0.36
    就好
    0.34
    coh
    0.32
     mediums
    0.31
     floors
    0.31
    ations
    0.31
     sizes
    0.31
     durations
    0.31
     domains
    0.30
    Activations Density 1.674%

    No Known Activations