© 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. 47502
    Prev
    Next
    INDEX
    Explanations

    key

    np_max-act · gemini-2.0-flash

    The neuron fires on mentions of dictionary “keys” (e.g. the token “keys” in code or its Spanish form “claves”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    key-value pairs in ascending and descending order in Python dictionaries.

    oai_token-act-pair · gpt-4o-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
     compositions
    -0.07
    pany
    -0.07
     zbo
    -0.07
    46
    -0.06
    weeney
    -0.06
    ulance
    -0.06
    Paper
    -0.06
    enarios
    -0.06
     Aquarium
    -0.06
    879
    -0.06
    POSITIVE LOGITS
     uğra
    0.06
    	elsif
    0.06
    exclude
    0.06
     magical
    0.06
    toLowerCase
    0.06
    	startActivity
    0.06
     Navigator
    0.06
     вій
    0.06
    .setdefault
    0.06
     nhánh
    0.06
    Activations Density 0.012%

    No Known Activations