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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 57616
    Prev
    Next
    INDEX
    Explanations

    Printing to console

    np_max-act · gemini-2.0-flash

    This neuron detects error‐handling print statements that output an “invalid input” or similar user‐facing error message.

    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
    vertex
    -0.07
     nexus
    -0.07
     promote
    -0.07
     landmark
    -0.07
     shuffle
    -0.07
    URL
    -0.07
     smaller
    -0.07
    -offset
    -0.07
     interactions
    -0.07
     polished
    -0.06
    POSITIVE LOGITS
    마사지
    0.06
    病院
    0.06
    组
    0.06
     cyclist
    0.06
    rement
    0.06
     recib
    0.06
     соблюд
    0.06
     entrar
    0.06
    -changing
    0.05
    iếm
    0.05
    Activations Density 0.008%

    No Known Activations