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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 90971
    Prev
    Next
    INDEX
    Explanations

    mention, said

    np_max-act · gemini-2.0-flash

    The neuron fires on meta‐commentary phrases that reference or pick up on what the other person said (e.g. “you said,” “you mentioned,” “as mentioned above”).

    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
     Print
    -0.07
     Manip
    -0.07
     awake
    -0.06
     puppy
    -0.06
     kit
    -0.06
     photographs
    -0.06
     screenshot
    -0.06
     vot
    -0.06
    ATEGORY
    -0.06
    -big
    -0.06
    POSITIVE LOGITS
    _MASTER
    0.07
     самое
    0.07
    _Anim
    0.07
     ию
    0.07
     Milli
    0.07
     topl
    0.06
     AssertionError
    0.06
     kadar
    0.06
    emez
    0.06
    дина
    0.06
    Activations Density 0.027%

    No Known Activations