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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 88775
    Prev
    Next
    INDEX
    Explanations

    amazing

    np_max-act · gemini-2.0-flash

    The neuron fires on self-referential, first-person expressions (e.g. “I,” “me,” “my,” “learning,” “around me”).

    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
    _Normal
    -0.07
    ερό
    -0.07
     stru
    -0.06
    (sample
    -0.06
     بهتر
    -0.06
    まり
    -0.06
    úi
    -0.06
    igmatic
    -0.06
     Gra
    -0.06
     winters
    -0.06
    POSITIVE LOGITS
    _TRANS
    0.07
    implode
    0.06
    ','=',$
    0.06
     Binder
    0.06
     TextStyle
    0.06
    Mapped
    0.06
    acionales
    0.06
     apparently
    0.06
    ίκη
    0.06
     nostalg
    0.06
    Activations Density 0.102%

    No Known Activations