© 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. 9505
    Prev
    Next
    INDEX
    Explanations

    Code and file paths

    np_max-act · gemini-2.0-flash

    The neuron is activated by code‐style or technical tokens—things like configuration directives, file‐path segments, and other programming‐style identifiers.

    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
     MET
    -0.06
     immigration
    -0.06
    AAC
    -0.06
     Koch
    -0.06
     FF
    -0.06
    .#
    -0.06
     hare
    -0.06
    pression
    -0.06
     intercepted
    -0.06
     Ra
    -0.06
    POSITIVE LOGITS
    .ToDouble
    0.07
    ตรว
    0.07
     演
    0.07
    [Any
    0.07
    вают
    0.07
    =lambda
    0.07
    422
    0.07
    ско
    0.06
    042
    0.06
     základě
    0.06
    Activations Density 0.004%

    No Known Activations