© 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. 59297
    Prev
    Next
    INDEX
    Explanations

    determined

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on “-ed” forms—i.e. past-tense or past-participle words (like determined, driven, devoted).

    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
     Muslims
    -0.06
     wrinkles
    -0.06
     плани
    -0.06
    �
    -0.06
     Transformers
    -0.06
    (subject
    -0.06
    (single
    -0.06
    .Single
    -0.06
    ーの
    -0.06
    reeNode
    -0.06
    POSITIVE LOGITS
     جان
    0.08
    egin
    0.07
    кс
    0.07
    .activ
    0.07
     μέρος
    0.07
     цел
    0.06
     GAL
    0.06
     både
    0.06
    lcd
    0.06
    ัพ
    0.06
    Activations Density 0.527%

    No Known Activations