© 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. 30914
    Prev
    Next
    INDEX
    Explanations

    is

    np_max-act · gemini-2.0-flash

    This neuron activates on the special “header ID” and role‐label markers (e.g. <|start_header_id|>, “assistant,” etc.) that appear in the conversation metadata rather than on normal text.

    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
    illez
    -0.07
     Guards
    -0.07
    .Output
    -0.06
     Battles
    -0.06
     اجر
    -0.06
     Soccer
    -0.06
    ้นท
    -0.06
    raham
    -0.06
     accomplishment
    -0.06
    ynec
    -0.06
    POSITIVE LOGITS
    _PD
    0.07
     här
    0.07
     annoying
    0.06
     Important
    0.06
    faction
    0.06
     ald
    0.06
     Ac
    0.06
    (hostname
    0.06
    @n
    0.06
     stál
    0.06
    Activations Density 0.086%

    No Known Activations