© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 128243
    Prev
    Next
    INDEX
    Explanations

    modifications or alterations

    np_max-act · gemini-2.0-flash

    This neuron activates on words and phrases from the framing or instruction of the task—terms like “contextualization,” “identify,” “information,” “your,” and other prompt‐setting language.

    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
     replen
    -0.07
    Bay
    -0.07
    Luckily
    -0.06
    Slot
    -0.06
    Whats
    -0.06
     rejected
    -0.06
    middle
    -0.06
    Ky
    -0.06
    Wizard
    -0.06
     jumps
    -0.06
    POSITIVE LOGITS
     ceased
    0.07
     ℝ
    0.07
    ाजन
    0.06
     Voyage
    0.06
    に見
    0.06
    	common
    0.06
     سلام
    0.06
    폐
    0.06
     inadequate
    0.06
    _APP
    0.06
    Activations Density 0.002%

    No Known Activations