© 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. 56240
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    The neuron is primarily detecting the infinitive “to” that introduces purpose or action (the “to ” in “to do something”).

    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
     Mast
    -0.06
     Crom
    -0.06
    .CreateTable
    -0.06
    -ball
    -0.06
     deutschland
    -0.06
     Sherlock
    -0.06
     bumps
    -0.06
     presup
    -0.06
     "(
    -0.06
     RATE
    -0.06
    POSITIVE LOGITS
    Backing
    0.07
     Replacement
    0.07
     resett
    0.06
     reforms
    0.06
    removeClass
    0.06
    0.06
    ตน
    0.06
     engaged
    0.06
    creating
    0.06
     ماي
    0.06
    Activations Density 0.059%

    No Known Activations