© 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. 57720
    Prev
    Next
    INDEX
    Explanations

    Cloud computing configurations

    np_max-act · gemini-2.0-flash

    The neuron fires on the special system/chat header and segment-boundary tokens (e.g. `<|start_header_id|>`, `<|end_header_id|>`, and associated markers) that delimit conversation turns.

    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
    .sprites
    -0.07
    _expiry
    -0.06
    	log
    -0.06
     đau
    -0.06
     ख
    -0.06
    wyn
    -0.06
     weakSelf
    -0.06
    quez
    -0.06
    ayload
    -0.06
    VERIFY
    -0.06
    POSITIVE LOGITS
     fireworks
    0.07
     prostitution
    0.07
     groupe
    0.06
     serviço
    0.06
    irtschaft
    0.06
     acompanh
    0.06
    leneck
    0.06
    correct
    0.06
     figur
    0.06
     holes
    0.06
    Activations Density 0.030%

    No Known Activations