© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Qwen2.5-7B-IT
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 12447
    Prev
    Next
    INDEX
    Explanations

    expressing belief

    np_max-act · gemini-2.0-flash

    Sentences or phrases expressing first-person statements of belief, certainty, or self-reference (e.g., "I", "I think", "I'm fairly certain", "I believe").

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     nederland
    -0.08
    ())/
    -0.07
    .Yes
    -0.07
    fter
    -0.07
    rieb
    -0.07
    普
    -0.07
    �
    -0.07
    Locked
    -0.07
    ichten
    -0.07
     dtype
    -0.07
    POSITIVE LOGITS
    annya
    0.08
    קור
    0.07
    (Stream
    0.07
     Canucks
    0.06
     căn
    0.06
    URES
    0.06
    romatic
    0.06
     בינ
    0.06
    Ѱ
    0.06
    .csrf
    0.06
    Activations Density 0.064%

    No Known Activations