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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 128586
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron primarily detects the starts of sentences or clauses—that is, tokens appearing just after punctuation or line breaks indicating a new sentence.

    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
     Locke
    -0.07
     듯
    -0.06
     transported
    -0.06
     SIDE
    -0.06
     Decoration
    -0.06
    โครงการ
    -0.06
    alette
    -0.06
    (phase
    -0.06
    TTY
    -0.06
    具体
    -0.06
    POSITIVE LOGITS
    aj
    0.07
    ハイ
    0.06
    <Date
    0.06
    AJ
    0.06
    (priv
    0.06
    [item
    0.06
     numerical
    0.06
    agher
    0.06
     říj
    0.06
     staffer
    0.06
    Activations Density 0.186%

    No Known Activations