© 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. 74050
    Prev
    Next
    INDEX
    Explanations

    code and programming

    np_max-act · gemini-2.0-flash

    The neuron fires on instruction-related tokens—especially numeric specifications and constraint words (e.g. “only,” “important,” “identify,” “should”) in the prompt.

    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
    _Q
    -0.08
    amb
    -0.06
    /pages
    -0.06
    oplay
    -0.06
    られる
    -0.06
    (sq
    -0.06
    ore
    -0.06
     apk
    -0.06
    contracts
    -0.06
    lags
    -0.06
    POSITIVE LOGITS
    」的
    0.08
     wrongly
    0.07
     Luke
    0.07
    周
    0.07
    ->{$
    0.07
     historical
    0.07
     preventing
    0.07
     ApplicationController
    0.06
     domácí
    0.06
     близько
    0.06
    Activations Density 0.002%

    No Known Activations