Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Qwen2.5-7B-IT
    4. Resid Post - 131k
    5. 19-RESID-POST-AA
    6. 87751
    Prev
    Next
    INDEX
    Explanations

    say code comments

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_19/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.19.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    .Util
    -0.08
     glfw
    -0.07
    RULE
    -0.07
     ngh
    -0.07
     balk
    -0.07
    꾹
    -0.07
    옛
    -0.07
     waterfront
    -0.06
    .training
    -0.06
    뷁
    -0.06
    POSITIVE LOGITS
     economía
    0.07
    �
    0.07
     Global
    0.07
     sperm
    0.07
    卫
    0.07
    oram
    0.06
     sms
    0.06
     rehab
    0.06
     })↵
    0.06
     sergeant
    0.06
    Activations Density 0.001%

    No Known Activations