© 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. 37968
    Prev
    Next
    INDEX
    Explanations

    disclaimers and advice

    np_max-act · gemini-2.0-flash

    legal disclaimers and warnings regarding financial and investment advice.

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

    The neuron detects disclaimer language, especially “no financial (or investment) advice”–style legal disclaimers.

    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
    normally
    -0.08
    @"
    -0.06
    regar
    -0.06
     curated
    -0.06
     rr
    -0.06
     ask
    -0.06
    _path
    -0.06
    _Pl
    -0.06
    -0.06
     Coff
    -0.06
    POSITIVE LOGITS
    âl
    0.08
     χω
    0.07
     станет
    0.06
    스의
    0.06
     พฤษภาคม
    0.06
    ๒
    0.06
    ницип
    0.06
     الذي
    0.06
     yaşan
    0.06
    _stride
    0.06
    Activations Density 0.006%

    No Known Activations