© 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. 124727
    Prev
    Next
    INDEX
    Explanations

    crypt

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of the prefix “crypt” in words (e.g. cryptorchidism, cryptogenic, cryptosporidiosis).

    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
    LEM
    -0.07
     уси
    -0.07
    652
    -0.07
     Adam
    -0.07
     Helper
    -0.07
     DLC
    -0.07
     Harrison
    -0.07
    LE
    -0.07
     swelling
    -0.07
    .L
    -0.07
    POSITIVE LOGITS
     crypt
    0.11
     Crypt
    0.11
    crypt
    0.09
    Crypt
    0.09
     vault
    0.08
     Vaults
    0.08
     Cry
    0.07
     '[
    0.07
    PropertyValue
    0.07
    không
    0.06
    Activations Density 0.003%

    No Known Activations