© 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. 89596
    Prev
    Next
    INDEX
    Explanations

    unction

    np_max-act · gemini-2.0-flash

    The neuron activates on words featuring the “tri‐” or “trio/ triple” root, i.e. terms denoting groups of three.

    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
    16
    -0.08
    12
    -0.08
    14
    -0.08
    17
    -0.07
    19
    -0.07
    24
    -0.07
    [iVar
    -0.06
    11
    -0.06
    13
    -0.06
    15
    -0.06
    POSITIVE LOGITS
    3
    0.09
     three
    0.09
     Three
    0.08
     iii
    0.07
     III
    0.07
     tripod
    0.07
    .Person
    0.07
     trio
    0.07
     трьох
    0.07
     трех
    0.07
    Activations Density 0.334%

    No Known Activations