© 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. 58483
    Prev
    Next
    INDEX
    Explanations

    reps

    np_max-act · gemini-2.0-flash

    This neuron detects the special metadata tokens that mark the start of assistant response headers (e.g. “<|start_header_id|>”).

    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
    τρι
    -0.07
    between
    -0.06
    آ
    -0.06
     bones
    -0.06
    ーレ
    -0.06
    middlewares
    -0.06
    .condition
    -0.06
    Loading
    -0.06
     Translate
    -0.06
     buena
    -0.06
    POSITIVE LOGITS
    .setter
    0.07
    選手
    0.07
     Singap
    0.07
    �
    0.07
    	wp
    0.07
    REV
    0.06
    จำก
    0.06
    配
    0.06
     makeshift
    0.06
    .surname
    0.06
    Activations Density 0.006%

    No Known Activations