© 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. 54500
    Prev
    Next
    INDEX
    Explanations

    degrees

    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
    Not in Any Lists

    No Comments

    Negative Logits
    anta
    -0.26
    issant
    -0.26
     Gerald
    -0.25
     主
    -0.24
    ibal
    -0.24
    pector
    -0.23
    isser
    -0.23
    iesen
    -0.23
    Reader
    -0.23
    åıį
    -0.23
    POSITIVE LOGITS
     Degrees
    0.30
     degrees
    0.28
    ä¸ĩ家
    0.27
    çģ«çĥ§
    0.27
    complete
    0.26
    comb
    0.25
    manufact
    0.25
    çĤĴä½ľ
    0.25
    shake
    0.25
    Degrees
    0.25
    Activations Density 0.003%

    No Known Activations