© 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. 13677
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash

    This neuron activates on words describing a perilous dungeon environment (e.g., “dangerous,” “dungeon,” “traps”).

    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
     TH
    -0.08
     уж
    -0.07
    _utils
    -0.07
    คำ
    -0.07
     tst
    -0.07
    于是
    -0.06
    .words
    -0.06
    Handling
    -0.06
     مه
    -0.06
     �
    -0.06
    POSITIVE LOGITS
    licted
    0.07
    ger
    0.07
    icial
    0.06
    DRAM
    0.06
     Coco
    0.06
    ,"↵
    0.06
    ographer
    0.06
    cısı
    0.06
    ịnh
    0.06
    approved
    0.06
    Activations Density 0.042%

    No Known Activations