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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 9231
    Prev
    Next
    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    The neuron is chiefly activated by special control markers (e.g. the `<|eot_id|>` end-of-text tokens) rather than ordinary text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    tokens that mark document boundaries or special control/metadata (special control tokens).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
    =d
    -0.08
    b
    -0.07
    .variables
    -0.07
    hdl
    -0.07
    енко
    -0.07
    20
    -0.07
     Trent
    -0.07
    ético
    -0.07
     dusk
    -0.07
    0
    -0.07
    POSITIVE LOGITS
     sexist
    0.07
    )));
    0.06
    <pcl
    0.06
    ').'</
    0.06
    ]}</
    0.06
     làn
    0.06
    addOn
    0.06
     mainAxisAlignment
    0.06
     sexism
    0.06
     iş
    0.06
    Activations Density 0.428%

    No Known Activations