© 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. 89475
    Prev
    Next
    INDEX
    Explanations

    reporting/explanations

    np_max-act · gemini-2.0-flash

    The neuron is sensitive to negation and contradiction cues (e.g. “No,” “not,” “while,” “does not”), flagging factual inconsistencies.

    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
     courseId
    -0.07
     Rh
    -0.06
     UPPER
    -0.06
     caught
    -0.06
     refugees
    -0.06
     sø
    -0.06
    (filepath
    -0.06
     hips
    -0.06
     tragic
    -0.06
    Texto
    -0.06
    POSITIVE LOGITS
    .logo
    0.08
    ocrine
    0.07
    _company
    0.07
     этого
    0.06
    ault
    0.06
     Bottle
    0.06
    .usage
    0.06
    шей
    0.06
    tement
    0.06
    χει
    0.06
    Activations Density 0.013%

    No Known Activations