© 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. 112487
    Prev
    Next
    INDEX
    Explanations

    Article/document excerpts

    np_max-act · gemini-2.0-flash

    This neuron detects French-language (non-English) text—it activates when the document contains French words or French sentences.

    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
    瑞典
    -0.08
    rep
    -0.08
    隐
    -0.08
    instant
    -0.08
    swift
    -0.07
    -res
    -0.07
    卿
    -0.07
     Prof
    -0.07
    Geometry
    -0.07
     organism
    -0.07
    POSITIVE LOGITS
     adicion
    0.08
     milestones
    0.07
    쯧
    0.07
     actualizar
    0.07
     lineHeight
    0.07
    inge
    0.06
    /end
    0.06
    opic
    0.06
     pij
    0.06
    놋
    0.06
    Activations Density 0.230%

    No Known Activations