© 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
    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. 41499
    Prev
    Next
    INDEX
    Explanations

    Personal stories

    np_max-act · gemini-2.0-flash

    The neuron selectively lights up on subordinating conjunctions and other small discourse-marker words (e.g. “if,” “then,” “it”) that introduce or link clauses.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    tokens that occur at the start of sentences or paragraphs (sentence/paragraph beginnings).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
    .Required
    -0.07
    controller
    -0.07
     Loading
    -0.07
    olding
    -0.07
     sequences
    -0.07
    .Sp
    -0.06
    Follow
    -0.06
    	where
    -0.06
    quete
    -0.06
    	aux
    -0.06
    POSITIVE LOGITS
    番号
    0.06
     ー
    0.06
     nech
    0.06
     배
    0.06
    ulace
    0.06
     BAB
    0.06
     amazed
    0.06
    ازند
    0.06
     Nasıl
    0.06
    hk
    0.06
    Activations Density 0.220%

    No Known Activations