Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 16172
    Prev
    Next
    INDEX
    Explanations

    dates

    np_max-act · gemini-2.0-flash

    This neuron specializes in spotting tokens that are part of specific date mentions (like day, month, or year components) in historical or opening-date contexts.

    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
     dimensions
    -0.06
     underrated
    -0.06
    rotation
    -0.06
    (timer
    -0.06
     pumping
    -0.06
     �
    -0.06
    .numpy
    -0.06
    (pa
    -0.06
    .Alpha
    -0.06
    _GPS
    -0.06
    POSITIVE LOGITS
    .....↵↵
    0.07
     знать
    0.07
    ออ
    0.07
     battleground
    0.06
     distractions
    0.06
     DO
    0.06
    $m
    0.06
    ์ได
    0.06
     ROC
    0.06
    _result
    0.06
    Activations Density 0.018%

    No Known Activations