© 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. 128881
    Prev
    Next
    INDEX
    Explanations

    why

    np_max-act · gemini-2.0-flash

    The neuron fires on words that signal explanation or justification (e.g. “why,” “reasoning,” “giving”).

    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
    BootApplication
    -0.07
    .notice
    -0.07
     Modify
    -0.06
     mare
    -0.06
    ναν
    -0.06
    acios
    -0.06
     knife
    -0.06
    modity
    -0.06
     coz
    -0.06
    溫
    -0.06
    POSITIVE LOGITS
    sem
    0.08
     معن
    0.07
     Epid
    0.06
    ">↵↵
    0.06
    jug
    0.06
     mü
    0.06
     freeing
    0.06
     deutsch
    0.06
     chuyên
    0.06
     Filtering
    0.06
    Activations Density 0.159%

    No Known Activations