© 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. 22-TRANSCODER-HP
    7. 129395
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    äºĽ
    -0.25
    vre
    -0.25
    ucht
    -0.24
     met
    -0.24
    meet
    -0.24
    ä¸ĢäºĽ
    -0.24
    umer
    -0.23
     meet
    -0.23
    æĬ½
    -0.23
    iti
    -0.23
    POSITIVE LOGITS
    eway
    0.35
    çīĮ
    0.30
    ARK
    0.28
    idon
    0.27
    odb
    0.27
    çĶ±æѤ
    0.27
    æ£Ĥ
    0.27
    第ä¸ī代
    0.26
    elow
    0.25
    .ModelAdmin
    0.25
    Activations Density 0.010%

    No Known Activations