© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 126579
    Prev
    Next
    INDEX
    Explanations

    HTML

    np_max-act · gemini-2.0-flash

    The neuron fires on HTML markup tokens—i.e. it detects tag names and their delimiters (like `<button>`, `<script>`, `<body>`, etc.).

    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
     Khal
    -0.08
     Mrs
    -0.07
    ians
    -0.06
     sniff
    -0.06
    	op
    -0.06
     لأن
    -0.06
    vro
    -0.06
    ヾ
    -0.06
     AIS
    -0.06
     Menschen
    -0.06
    POSITIVE LOGITS
     ];↵
    0.06
    stead
    0.06
     disastr
    0.06
     Bình
    0.06
    _Execute
    0.06
    .Block
    0.06
    _DATA
    0.06
    (coord
    0.06
     Kent
    0.06
     Anything
    0.06
    Activations Density 0.040%

    No Known Activations