© 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. 7112
    Prev
    Next
    INDEX
    Explanations

    in front

    np_max-act · gemini-2.0-flash

    The neuron fires on words and phrases indicating leading position or movement ahead—e.g. “front‐running,” “front,” “advance,” or “ahead.”

    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
     stared
    -0.07
    484
    -0.06
    fit
    -0.06
    _sink
    -0.06
     Nature
    -0.06
     PM
    -0.06
     téléphone
    -0.06
     Leap
    -0.06
     cade
    -0.06
    Options
    -0.06
    POSITIVE LOGITS
     scouts
    0.07
    /apps
    0.07
     authentic
    0.07
     علاق
    0.07
     krás
    0.06
     ülk
    0.06
    κό
    0.06
    _skill
    0.06
     agility
    0.06
    (saved
    0.06
    Activations Density 0.014%

    No Known Activations