© 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. 111613
    Prev
    Next
    INDEX
    Explanations

    help

    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
    èģ
    -0.26
    ath
    -0.25
    åıijå±ķè¶ĭåĬ¿
    -0.25
    erd
    -0.24
    åĽ½å®¶éĩįçĤ¹
    -0.23
    CLE
    -0.23
     lit
    -0.23
    ä¸Ģ声
    -0.23
     occasion
    -0.23
    åľ¨ä»ĸçļĦ
    -0.22
    POSITIVE LOGITS
    abic
    0.28
     persuasive
    0.28
    icus
    0.27
     persuade
    0.25
    -factor
    0.25
     brace
    0.24
    平衡
    0.24
    _interfaces
    0.24
    /Branch
    0.24
    identification
    0.24
    Activations Density 0.005%

    No Known Activations