© 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. 42925
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    The neuron identifies passive‐voice verb constructions, i.e. forms of “be” followed by a past participle (e.g. “are designed,” “is produced,” “provide[d]”).

    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
    raw
    -0.07
    having
    -0.07
    616
    -0.06
    病
    -0.06
    (session
    -0.06
    here
    -0.06
     Lakes
    -0.06
    foundation
    -0.06
     undergraduate
    -0.06
     forests
    -0.06
    POSITIVE LOGITS
    .convert
    0.07
     '#
    0.07
     т
    0.06
    ","#
    0.06
    .dst
    0.06
     "#
    0.06
    当前
    0.06
    ,model
    0.06
    Inf
    0.06
    :UIControlStateNormal
    0.06
    Activations Density 0.029%

    No Known Activations