© 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. 57575
    Prev
    Next
    INDEX
    Explanations

    Negation

    np_max-act · gemini-2.0-flash

    The neuron activates on statements where the speaker (the assistant) denies having personal qualities or experiences—e.g. “I do not have a favorite color,” “I do not experience feelings,” 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
     rendre
    -0.06
    екту
    -0.06
    世界
    -0.06
    gba
    -0.06
    __(/*!
    -0.06
    Metal
    -0.06
    .eth
    -0.06
    Phrase
    -0.06
     Princess
    -0.06
     mimeType
    -0.06
    POSITIVE LOGITS
    crew
    0.07
     onions
    0.06
    zik
    0.06
    áfico
    0.06
     FP
    0.06
    OPT
    0.06
     erk
    0.06
     编辑
    0.06
     bfd
    0.06
    मन
    0.06
    Activations Density 0.028%

    No Known Activations