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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 6454
    Prev
    Next
    INDEX
    Explanations

    quality

    np_max-act · gemini-2.0-flash

    The neuron fires on descriptive phrases emphasizing product or service quality and reliability (e.g. “high-quality,” “reliable,” “quality control”).

    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
     wrongful
    -0.07
    ToBounds
    -0.06
    公園
    -0.06
     एक
    -0.06
    	ref
    -0.06
     Blind
    -0.06
    字符串
    -0.06
    _des
    -0.06
    RequestId
    -0.06
     illusion
    -0.06
    POSITIVE LOGITS
    ляд
    0.07
     IDEOGRAPH
    0.07
     modifies
    0.06
    Timestamp
    0.06
     Assy
    0.06
     Bütün
    0.06
     diren
    0.06
     değiştir
    0.06
     اجتماع
    0.06
     RAM
    0.06
    Activations Density 0.014%

    No Known Activations