© 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. 54780
    Prev
    Next
    INDEX
    Explanations

    name fragments ending in "am"

    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
    clas
    -0.26
    akens
    -0.24
    anni
    -0.23
    èĶ·
    -0.23
    pects
    -0.23
    numer
    -0.23
     неÑĤ
    -0.23
     buggy
    -0.23
    ivable
    -0.23
    andbox
    -0.23
    POSITIVE LOGITS
    arc
    0.32
    aria
    0.31
     arc
    0.30
    ike
    0.30
    ely
    0.29
    弧
    0.29
    it
    0.29
    oss
    0.28
     Arc
    0.28
    arie
    0.28
    Activations Density 0.014%

    No Known Activations