© 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. 6138
    Prev
    Next
    INDEX
    Explanations

    research papers

    np_max-act · gemini-2.0-flash

    The neuron is specialized in detecting in‐text scholarly citations or reference markers (e.g. author names, “et al.,” years, and other citation fragments).

    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
    ermann
    -0.08
     spacing
    -0.07
    emem
    -0.06
    ozo
    -0.06
     reload
    -0.06
    -eyed
    -0.06
     seper
    -0.06
     elapsedTime
    -0.06
    cede
    -0.06
    ч
    -0.06
    POSITIVE LOGITS
     بیمه
    0.07
    もしれない
    0.07
     ((__
    0.06
    odní
    0.06
     przy
    0.06
     ذ
    0.06
     bastante
    0.06
    إنجليزية
    0.06
    capitalize
    0.06
     CommonModule
    0.06
    Activations Density 0.024%

    No Known Activations