© 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. 56538
    Prev
    Next
    INDEX
    Explanations

    additions

    np_max-act · gemini-2.0-flash

    This neuron activates on words and short phrases that mark direct, blunt politeness strategies (e.g. “direct,” “honesty,” “reflect”) in tech‐style discourse.

    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
    Saint
    -0.07
    Phi
    -0.06
    raman
    -0.06
     cds
    -0.06
     pillow
    -0.06
    ामन
    -0.06
    287
    -0.06
    bih
    -0.06
     laps
    -0.06
     supporting
    -0.06
    POSITIVE LOGITS
     upro
    0.07
    시에
    0.07
    WS
    0.06
     zast
    0.06
     conex
    0.06
    え
    0.06
    attached
    0.06
    だって
    0.06
     ElementRef
    0.06
     ilişkin
    0.06
    Activations Density 0.315%

    No Known Activations