© 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. 13099
    Prev
    Next
    INDEX
    Explanations

    general text

    np_max-act · gemini-2.0-flash

    The neuron detects directive or policy‐specification language—that is, the instruction and guideline sections of the prompt where roles, rules, or required behaviors are laid out.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    financial calculations involving interest rates.

    oai_token-act-pair · gpt-4o-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
    :**
    -0.06
    .gstatic
    -0.06
    ocks
    -0.06
    zos
    -0.06
    .lv
    -0.06
    อลลาร
    -0.06
    icus
    -0.06
    лів
    -0.06
    .mybatisplus
    -0.06
    .Skin
    -0.06
    POSITIVE LOGITS
    trajectory
    0.07
     efficiencies
    0.06
    [J
    0.06
     Want
    0.06
     Statistical
    0.06
     './../../
    0.06
    ikipedia
    0.06
    ت
    0.06
     leng
    0.06
     realities
    0.06
    Activations Density 0.075%

    No Known Activations