© 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. 57711
    Prev
    Next
    INDEX
    Explanations

    j

    np_max-act · gemini-2.0-flash

    The neuron fires on code comments and statements that describe printing contiguous subarrays (e.g. “print the array from i to j”).

    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
    Prog
    -0.07
     Hag
    -0.07
     spectra
    -0.07
    Sm
    -0.07
     Julie
    -0.07
    具有
    -0.06
    boru
    -0.06
     Includes
    -0.06
     Lyrics
    -0.06
    ाएग
    -0.06
    POSITIVE LOGITS
     FAT
    0.06
     khách
    0.06
    	hit
    0.06
     LEN
    0.06
     Treasury
    0.06
    reserve
    0.06
    IU
    0.06
     offend
    0.06
     біль
    0.06
     nợ
    0.06
    Activations Density 0.269%

    No Known Activations