© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 130080
    Prev
    Next
    INDEX
    Explanations

    round

    np_max-act · gemini-2.0-flash

    The neuron is selectively triggered by the system‐ and turn‐boundary markers that frame a chat exchange (e.g. tokens delineating “begin_of_text,” “start_header_id,” “end_header_id,” or “eot_id”), i.e. it detects the special metadata tokens used to mark up the multi‐turn dialogue.

    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
    #+#+
    -0.08
    tatus
    -0.07
    _dense
    -0.07
     dime
    -0.07
    mtree
    -0.07
     oldest
    -0.06
     созда
    -0.06
    .desc
    -0.06
     told
    -0.06
    �
    -0.06
    POSITIVE LOGITS
    辛
    0.07
    .getInputStream
    0.07
    REA
    0.07
    熊
    0.07
    sand
    0.07
    ’l
    0.06
     اقتصاد
    0.06
    684
    0.06
    .Sql
    0.06
     IP
    0.06
    Activations Density 0.006%

    No Known Activations