© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 126331
    Prev
    Next
    INDEX
    Explanations

    option/answer

    np_max-act · gemini-2.0-flash

    The neuron activates on the word “option,” i.e. when the text presents the multiple‐choice prompt keyword option.

    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
     olmaktadır
    -0.08
     وكانت
    -0.06
    (av
    -0.06
    átek
    -0.06
    شم
    -0.06
     Пред
    -0.06
    inois
    -0.06
    قدر
    -0.06
    Knife
    -0.06
    }:{
    -0.06
    POSITIVE LOGITS
     restraint
    0.07
     jeu
    0.06
     죽
    0.06
    _flag
    0.06
     Enter
    0.05
     Haj
    0.05
    Enter
    0.05
    'ét
    0.05
    elsey
    0.05
    تدى
    0.05
    Activations Density 0.002%

    No Known Activations