© 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. 89181
    Prev
    Next
    INDEX
    Explanations

    single

    np_max-act · gemini-2.0-flash

    This neuron responds to words that emphasize a single, unified, or common solution or starting point (e.g., “single,” “unified,” “common,” “starting”).

    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
     twentieth
    -0.07
     exposition
    -0.07
    Tue
    -0.06
    ANG
    -0.06
    rors
    -0.06
    -modules
    -0.06
     shoulders
    -0.06
    078
    -0.06
     Guerr
    -0.06
    URRENCY
    -0.06
    POSITIVE LOGITS
    _disp
    0.08
    ItemSelectedListener
    0.07
    Stats
    0.07
    chyb
    0.06
     wir
    0.06
    stop
    0.06
     بررسی
    0.06
     consolid
    0.06
    ('.')↵
    0.06
     Persons
    0.06
    Activations Density 0.021%

    No Known Activations