© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 33-TRANSCODER-HP
    7. 53006
    Prev
    Next
    INDEX
    Explanations

    Lifting

    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
    çĽĺ
    -0.28
    æĭīå¼Ģ
    -0.28
    Shar
    -0.27
    æĭĴ
    -0.26
    ftar
    -0.26
    plate
    -0.26
    èĤ¡
    -0.26
     hubs
    -0.25
    JE
    -0.25
    ÑĢоÑģ
    -0.25
    POSITIVE LOGITS
     reach
    0.28
     reaches
    0.26
    reach
    0.26
    èIJ½åĪ°å®ŀå¤Ħ
    0.25
    know
    0.24
    è¾¾åĪ°
    0.24
    .setHeight
    0.24
     Pin
    0.24
     pin
    0.24
    è¶Ĭè¿ĩ
    0.24
    Activations Density 0.028%

    No Known Activations