© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 68380
    Prev
    Next
    INDEX
    Explanations

    Conversational writing

    np_max-act · gemini-2.0-flash

    The neuron responds to constructions that state something is missing or required—e.g. phrases like “the only things … missing are,” “needed,” “will be,” “putting,” and “are” that introduce what must or will be supplied.

    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
     locks
    -0.06
    ManagedObject
    -0.06
    (outputs
    -0.06
     Owens
    -0.06
    :Add
    -0.06
     απ
    -0.06
     D
    -0.06
    경
    -0.06
    peating
    -0.06
     dobré
    -0.05
    POSITIVE LOGITS
    heets
    0.07
     کودکان
    0.07
     RCC
    0.07
     dossier
    0.07
     tom
    0.07
     activ
    0.07
    Ik
    0.07
     Rams
    0.07
    FILE
    0.06
     tou
    0.06
    Activations Density 0.074%

    No Known Activations