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. 24425
    Prev
    Next
    INDEX
    Explanations

    riter

    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
    贩卖
    -0.07
    国土
    -0.07
    jualan
    -0.07
    四个方面
    -0.07
    ​​
    -0.06
     góc
    -0.06
     menj
    -0.06
    gaben
    -0.06
    车上
    -0.06
    gons
    -0.06
    POSITIVE LOGITS
     shuffle
    0.07
     subroutine
    0.07
     Washington
    0.07
     stab
    0.07
     stared
    0.07
    	continue
    0.07
    .ut
    0.07
     International
    0.07
     Initiative
    0.07
    笑笑
    0.06
    Activations Density 0.000%

    No Known Activations