© 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. 57776
    Prev
    Next
    INDEX
    Explanations

    Code and errors

    np_max-act · gemini-2.0-flash

    The neuron flags lines that look like Java exception stack‐trace entries (i.e. the “at … (…)” call‐stack lines).

    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
    _DAT
    -0.07
     bien
    -0.07
     cushions
    -0.06
    ाप
    -0.06
    _TRI
    -0.06
    ា�
    -0.06
     trọng
    -0.06
     discomfort
    -0.06
    modulo
    -0.06
     Wi
    -0.06
    POSITIVE LOGITS
     ／:
    0.07
    ／:
    0.06
    BinContent
    0.06
    bine
    0.06
     French
    0.06
     Tango
    0.06
     TAKE
    0.06
    gos
    0.06
    developer
    0.06
    _REGS
    0.06
    Activations Density 0.006%

    No Known Activations