© 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. 41197
    Prev
    Next
    INDEX
    Explanations

    is

    np_max-act · gemini-2.0-flash

    The neuron fires on the project‐name token in GNU license headers of the form “[ProjectName] is free software,” i.e. it detects that boilerplate “is free software” declaration.

    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
    _mpi
    -0.06
     Preserve
    -0.06
    _calibration
    -0.06
    	cc
    -0.06
    .sim
    -0.06
     Forbidden
    -0.06
    (ent
    -0.06
    irma
    -0.06
     failure
    -0.06
    Pages
    -0.06
    POSITIVE LOGITS
    (INT
    0.06
     zeigt
    0.06
    							↵
    0.06
     HSV
    0.06
     ethos
    0.06
     IMS
    0.06
    ()");↵
    0.06
    uciones
    0.06
    .hostname
    0.06
    voří
    0.06
    Activations Density 0.004%

    No Known Activations