© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 59429
    Prev
    Next
    INDEX
    Explanations

    cautionary language

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens in cautionary or disclaimer statements—especially around urging verification or noting limitations.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    該
    -0.06
    ppo
    -0.06
    ε
    -0.06
    ウェ
    -0.06
    Allocator
    -0.06
     UCHAR
    -0.06
     Contractor
    -0.06
    커
    -0.06
     Kuala
    -0.06
     awkward
    -0.05
    POSITIVE LOGITS
     setIs
    0.07
     SCORE
    0.07
    .nome
    0.07
     Bake
    0.07
    (fill
    0.07
     eyes
    0.06
    iben
    0.06
    <Props
    0.06
    taken
    0.06
    *size
    0.06
    Activations Density 0.038%

    No Known Activations