© 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. 112192
    Prev
    Next
    INDEX
    Explanations

    Educational background

    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
     learning
    -0.43
     proficiency
    -0.40
     Skill
    -0.38
    åŁ¹è®Ń
    -0.38
     skill
    -0.38
    æĬĢèĥ½
    -0.37
    skill
    -0.36
    å¸ĪèµĦ
    -0.35
    skills
    -0.35
     Mastery
    -0.35
    POSITIVE LOGITS
    æµ£
    0.27
    egend
    0.27
    AGER
    0.25
     coll
    0.25
     Kle
    0.24
    ottes
    0.24
    èĽ²
    0.24
    ç¦ĭ
    0.23
    iden
    0.23
     teen
    0.23
    Activations Density 0.001%

    No Known Activations