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. 262
    Prev
    Next
    INDEX
    Explanations

    chest

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on mentions of the chest area (e.g., “chest”) in physical descriptions.

    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
    _allocation
    -0.09
     distribution
    -0.07
    [op
    -0.07
    _temperature
    -0.07
     remake
    -0.07
    524
    -0.07
    illi
    -0.07
    798
    -0.07
     linkage
    -0.07
    osal
    -0.07
    POSITIVE LOGITS
     Chest
    0.13
     chest
    0.12
     chests
    0.11
    edics
    0.08
     Ethan
    0.07
     nestled
    0.07
    ><?=$
    0.07
     should
    0.07
    >{!!
    0.07
    cab
    0.07
    Activations Density 0.006%

    No Known Activations