© 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. 19-RESID-POST-AA
    11. 111683
    Prev
    Next
    INDEX
    Explanations

    be

    np_max-act · gemini-2.0-flash

    the neuron detects authorial/explanatory meta-discourse — sentences that give explanations, guidance, examples, or commentary about the text (e.g., translator/reader instructions and descriptive remarks).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31

    be

    np_max-act-logits · gpt-5-miniTriggered by @yooniel31

    hedging language Method used: 2 — detects hedging language

    np_max-act · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_19/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.19.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    .removeClass
    -0.06
     섬
    -0.06
     dévelop
    -0.06
    	endif
    -0.06
    資料
    -0.06
    fluence
    -0.06
    ALLED
    -0.06
     resist
    -0.06
     สพ
    -0.06
    roz
    -0.06
    POSITIVE LOGITS
     każ
    0.07
     Moses
    0.06
     Boehner
    0.06
     immunity
    0.06
    niční
    0.06
     Shoes
    0.06
    677
    0.06
     recognise
    0.06
     shel
    0.06
    076
    0.06
    Activations Density 0.199%

    No Known Activations