Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 129503
    Prev
    Next
    INDEX
    Explanations

    file

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of writing to or using “file” (e.g. logging to a file or storing data in a file).

    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
     servisi
    -0.07
     '!
    -0.07
    bcrypt
    -0.06
    掛
    -0.06
    	define
    -0.06
    urst
    -0.06
    IZE
    -0.06
    ials
    -0.06
     marginBottom
    -0.06
     domingo
    -0.06
    POSITIVE LOGITS
     local
    0.07
     реак
    0.07
    ':"
    0.07
    τας
    0.06
     ensued
    0.06
    _PROJECT
    0.06
    pute
    0.06
    inality
    0.06
    Schedulers
    0.06
     Proof
    0.06
    Activations Density 0.015%

    No Known Activations