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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 58986
    Prev
    Next
    INDEX
    Explanations

    prepositions/conjunctions

    np_max-act · gemini-2.0-flash

    The neuron activates on English prepositions and other small function words that signal relationships (e.g. in, of, to, by, with, from, towards).

    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
    Você
    -0.07
    axon
    -0.07
     dazu
    -0.07
     Thomson
    -0.07
    zbek
    -0.06
    _destination
    -0.06
    orpion
    -0.06
     ganze
    -0.06
     inde
    -0.06
     twelve
    -0.06
    POSITIVE LOGITS
    .forward
    0.07
    '=>"
    0.06
    etect
    0.06
    entic
    0.06
     upon
    0.06
    .predict
    0.06
    (Of
    0.06
    吉
    0.06
    .gz
    0.06
    QUERY
    0.06
    Activations Density 0.212%

    No Known Activations