© 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. 57618
    Prev
    Next
    INDEX
    Explanations

    must

    np_max-act · gemini-2.0-flash

    grammatical errors and clarity issues in written text.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron mainly detects descriptions of personal opinions or stances (phrases where the author states “I think,” “I believe,” or otherwise expresses their viewpoint).

    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
    ega
    -0.08
    ISM
    -0.07
    بير
    -0.07
     soils
    -0.07
    ení
    -0.07
     ACTIONS
    -0.07
    ümü
    -0.06
    (mon
    -0.06
    ební
    -0.06
    binary
    -0.06
    POSITIVE LOGITS
     Ike
    0.06
    .awtextra
    0.06
    .setWindowTitle
    0.06
    ;?#
    0.06
     têm
    0.06
     eid
    0.06
     zayıf
    0.06
    itoris
    0.06
     li
    0.06
    'name
    0.06
    Activations Density 0.043%

    No Known Activations