© 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 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. 57184
    Prev
    Next
    INDEX
    Explanations

    awks

    np_max-act · gemini-2.0-flash

    The neuron responds to emotionally or morally charged content words—terms like “heart,” “love,” “faithless,” “conscience,” “serpent,” etc.—that carry moral or emotive weight.

    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
     рус
    -0.07
     سنت
    -0.06
    这是
    -0.06
     Firewall
    -0.06
    努
    -0.06
    -tree
    -0.06
    six
    -0.06
    ]=="
    -0.06
    莱
    -0.06
    直
    -0.06
    POSITIVE LOGITS
    textField
    0.07
    .ImageLayout
    0.06
     باشگاه
    0.06
     ACS
    0.06
    лара
    0.06
    CCI
    0.06
    AGED
    0.06
    anging
    0.06
    .raises
    0.06
    NC
    0.05
    Activations Density 0.083%

    No Known Activations