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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 56006
    Prev
    Next
    INDEX
    Explanations

    teaching/explaining

    np_max-act · gemini-2.0-flash

    This neuron detects the Japanese polite request construction “～について教えてください” (the phrase “about X, please tell me”).

    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
    格
    -0.07
    /The
    -0.06
    �
    -0.06
    	module
    -0.06
     ειδ
    -0.06
    ани
    -0.06
    .mass
    -0.06
     Hag
    -0.06
    adí
    -0.06
     dazzling
    -0.06
    POSITIVE LOGITS
    wifi
    0.07
    udes
    0.06
    loha
    0.06
     RESP
    0.06
     preview
    0.06
    *)"
    0.06
    iciální
    0.06
     lesb
    0.06
     rainbow
    0.06
    clazz
    0.06
    Activations Density 0.015%

    No Known Activations