© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 57646
    Prev
    Next
    INDEX
    Explanations

    BY

    np_max-act · gemini-2.0-flash

    identifiers and components related to the system calls in programming code.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron detects the uppercase boilerplate comment that a file is auto-generated (e.g. “THIS FILE IS GENERATED … DO NOT EDIT.”).

    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
     yếu
    -0.06
    	K
    -0.06
    _started
    -0.06
     sunset
    -0.06
    ]."
    -0.06
    _reads
    -0.06
    Warn
    -0.06
    ्रह
    -0.06
     fermentation
    -0.06
    成為
    -0.06
    POSITIVE LOGITS
    rench
    0.07
    (Scene
    0.07
    maxcdn
    0.07
    .GroupBox
    0.07
    _guess
    0.06
     promised
    0.06
    former
    0.06
    (cl
    0.06
     unb
    0.06
     implic
    0.06
    Activations Density 0.001%

    No Known Activations