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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 39152
    Prev
    Next
    INDEX
    Explanations

    database queries

    np_max-act · gemini-2.0-flash

    The neuron detects code‐block fence markers (backtick sequences) indicating the start or end of a code snippet.

    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
     baskets
    -0.07
    ר
    -0.07
    .car
    -0.07
    und
    -0.06
     Targets
    -0.06
     mt
    -0.06
    _installed
    -0.06
    貌
    -0.06
     왕
    -0.06
     graphs
    -0.06
    POSITIVE LOGITS
     Tem
    0.07
    ("#{
    0.06
    ải
    0.06
    _COMPARE
    0.06
     simply
    0.06
    .FLOAT
    0.06
    	import
    0.06
     stringWith
    0.06
    .trigger
    0.06
    :maj
    0.06
    Activations Density 0.021%

    No Known Activations