© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 126563
    Prev
    Next
    INDEX
    Explanations

    Descriptions of locations/events

    np_max-act · gemini-2.0-flash

    The neuron responds to tokens that signal past roles or biographical history (e.g. “formerly,” “previously,” “from,” “to”).

    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
    hed
    -0.07
    vided
    -0.06
    น
    -0.06
    Playlist
    -0.06
     conformity
    -0.06
    ensation
    -0.06
    %%%%%%%%%%%%%%%%
    -0.06
     anterior
    -0.06
     haunted
    -0.06
     progressed
    -0.06
    POSITIVE LOGITS
     alleges
    0.08
     помещ
    0.07
    되었습니다
    0.07
     �
    0.07
    ("//*[@
    0.07
    tım
    0.07
     Nicht
    0.07
    ़्
    0.07
     ÜNİVERSİTESİ
    0.07
    (parseInt
    0.06
    Activations Density 0.157%

    No Known Activations