Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 127624
    Prev
    Next
    INDEX
    Explanations

    /*

    np_max-act · gemini-2.0-flash

    This neuron selectively fires on code‐comment annotations—particularly the “@” markers inside C-style comment blocks (e.g. Javadoc or annotation comments).

    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
     Becky
    -0.07
     gelenek
    -0.07
     Diesel
    -0.07
     ple
    -0.07
    eng
    -0.07
     erle
    -0.07
    收益
    -0.07
    篇
    -0.06
     پژوهش
    -0.06
    Y
    -0.06
    POSITIVE LOGITS
     /*
    0.10
    ;/*
    0.07
    “When
    0.07
     Abb
    0.07
    aic
    0.07
    /*
    0.07
     chap
    0.07
     Č
    0.07
     Loop
    0.06
     Santa
    0.06
    Activations Density 0.009%

    No Known Activations