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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 2358
    Prev
    Next
    INDEX
    Explanations

    turned on

    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
    _before
    -0.07
     �
    -0.07
     בדי
    -0.07
     Negot
    -0.07
    nev
    -0.07
    )]↵
    -0.07
     sprzeda
    -0.07
    chen
    -0.07
    memcmp
    -0.07
    揖
    -0.07
    POSITIVE LOGITS
    almart
    0.07
    Language
    0.07
    Routes
    0.07
    尤其是
    0.07
    /oct
    0.06
     logic
    0.06
    呼吸道
    0.06
    inicio
    0.06
    _ABC
    0.06
    _TOTAL
    0.06
    Activations Density 0.050%

    No Known Activations