© 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
    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. 13664
    Prev
    Next
    INDEX
    Explanations

    prefixes

    np_max-act · gemini-2.0-flash

    This neuron consistently activates on the subword “inter” (as in inter-religious, inter-cultural, etc.).

    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
     frase
    -0.07
     一般
    -0.07
    علومات
    -0.06
     weave
    -0.06
     regex
    -0.06
    一般
    -0.06
     бан
    -0.06
    (ab
    -0.06
     padr
    -0.06
     Kadın
    -0.06
    POSITIVE LOGITS
     impending
    0.07
    ộp
    0.06
    嫌
    0.06
     restitution
    0.06
     Anyone
    0.06
     Mount
    0.06
    buster
    0.06
    achen
    0.06
    iger
    0.06
    mal
    0.06
    Activations Density 0.078%

    No Known Activations