© 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. 52882
    Prev
    Next
    INDEX
    Explanations

    questions and answers

    np_max-act · gemini-2.0-flash

    It detects the start of assistant replies / tokens marking an assistant speaker turn in the conversation.

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    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
    uffs
    -0.06
    841
    -0.06
    -focus
    -0.06
     captured
    -0.06
     GIS
    -0.06
    vault
    -0.06
    _heat
    -0.06
    techn
    -0.06
    resolution
    -0.06
     Caption
    -0.06
    POSITIVE LOGITS
     yaptığ
    0.08
     лекар
    0.07
     peny
    0.07
     Pang
    0.06
     Fucked
    0.06
     místě
    0.06
    śmy
    0.06
    .firstName
    0.06
    ümüzde
    0.06
    ุงเทพ
    0.06
    Activations Density 0.107%

    No Known Activations