© 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. 45781
    Prev
    Next
    INDEX
    Explanations

    collect

    np_max-act · gemini-2.0-flash

    sentences detailing the collection and handling of personal data in privacy policies.

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

    The neuron flags language describing the collection or handling of personal data (e.g. “collect,” “information,” “personal,” “data”) in privacy‐policy–style text.

    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
     xấu
    -0.06
    ствен
    -0.06
     Geile
    -0.06
    words
    -0.06
     свои
    -0.06
     forme
    -0.06
    ปล
    -0.06
    .vocab
    -0.06
     ZERO
    -0.06
     lực
    -0.06
    POSITIVE LOGITS
    ータ
    0.07
     Obt
    0.07
     important
    0.07
     Represent
    0.06
     Проф
    0.06
     commonly
    0.06
    .He
    0.06
    _ph
    0.06
    。她
    0.06
     talep
    0.06
    Activations Density 0.009%

    No Known Activations