© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 41325
    Prev
    Next
    INDEX
    Explanations

    ,

    np_max-act · gemini-2.0-flash

    This neuron fires on mentions of specific file types (e.g. “PDF file”) in technical instructions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    　　 　 　
    -0.07
    价格
    -0.06
    /WebAPI
    -0.06
    Exercise
    -0.06
     Glory
    -0.06
    redient
    -0.06
    Desktop
    -0.06
    amiliar
    -0.06
    Transactional
    -0.06
     Politics
    -0.06
    POSITIVE LOGITS
    �
    0.07
    cka
    0.07
     المن
    0.06
    ЕС
    0.06
     Pek
    0.06
    Dos
    0.06
     Cul
    0.06
    iyon
    0.06
    τον
    0.06
     repos
    0.06
    Activations Density 0.087%

    No Known Activations