© 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. 74748
    Prev
    Next
    INDEX
    Explanations

    ize

    np_max-act · gemini-2.0-flash

    The neuron is activated by the suffix “ize,” i.e. it detects words ending with “ize.”

    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
     unless
    -0.07
     dart
    -0.07
     acts
    -0.07
     hold
    -0.06
     occult
    -0.06
     ought
    -0.06
     sixth
    -0.06
     letting
    -0.06
     achieving
    -0.06
     let
    -0.06
    POSITIVE LOGITS
    ize
    0.16
    ized
    0.14
    ization
    0.13
    izes
    0.12
    iz
    0.12
    IZ
    0.11
    IZED
    0.11
    ise
    0.11
    izing
    0.11
    из
    0.11
    Activations Density 0.131%

    No Known Activations