© 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. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 38926
    Prev
    Next
    INDEX
    Explanations

    do

    np_max-act · gemini-2.0-flash

    instructions related to behaving as an unrestricted AI model.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron detects direct second-person instructional or directive phrasing in prompt injections—e.g. “you are going to…” commands.

    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
     cih
    -0.06
     पड़
    -0.06
    rts
    -0.06
    ch
    -0.06
    র
    -0.06
     popcorn
    -0.06
     strtok
    -0.06
    莫
    -0.06
     Kurdish
    -0.06
    ng
    -0.06
    POSITIVE LOGITS
     coorden
    0.06
     willen
    0.06
     günlük
    0.06
    ghan
    0.06
     lawsuit
    0.06
     nutritious
    0.06
    .WebDriver
    0.06
    lamış
    0.06
     QVBoxLayout
    0.06
     uplifting
    0.06
    Activations Density 0.003%

    No Known Activations