© 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. 59165
    Prev
    Next
    INDEX
    Explanations

    Last

    np_max-act · gemini-2.0-flash

    The neuron detects the word “Last” when it begins a time reference (as in “Last Thursday,” “Last week,” etc.).

    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
    ویزی
    -0.08
     PLUGIN
    -0.08
     Peng
    -0.07
    biên
    -0.07
    berg
    -0.07
    >().
    -0.07
     포
    -0.07
    _encoder
    -0.07
     zví
    -0.07
     keyboardType
    -0.07
    POSITIVE LOGITS
     last
    0.16
     Last
    0.15
    Last
    0.14
    .last
    0.14
    last
    0.13
    .Last
    0.13
     LAST
    0.12
    (last
    0.12
    	last
    0.11
    ,last
    0.10
    Activations Density 0.034%

    No Known Activations