© 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. 126716
    Prev
    Next
    INDEX
    Explanations

    forum posts

    np_max-act · gemini-2.0-flash

    This neuron fires on the special boundary token (the newline after the <|end_header_id|>) that marks the start of the assistant’s reply.

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

    Whether the text is part of the model/assistant's reply (assistant-generated content / assistant turn indicator).

    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
     Lawson
    -0.07
    isbn
    -0.06
     illustrations
    -0.06
    Child
    -0.06
    ени
    -0.06
    +'\
    -0.06
     disciples
    -0.06
    yh
    -0.06
    nop
    -0.06
     разреш
    -0.06
    POSITIVE LOGITS
     erectile
    0.06
     overhead
    0.06
    [layer
    0.06
    timing
    0.06
     Metallic
    0.06
     Sociology
    0.06
    ,n
    0.06
     بازیگر
    0.06
    ไว
    0.06
    (Token
    0.06
    Activations Density 0.040%

    No Known Activations