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. 42346
    Prev
    Next
    INDEX
    Explanations

    code logs

    np_max-act · gemini-2.0-flash

    This neuron activates on standalone code identifiers or component names—especially CamelCase or ALL-CAP words like Reshape, OP_BasicMovement, ExtractBitsBlock, Logger, HTTP, SetVersion, etc.

    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
     Rescue
    -0.07
    (ld
    -0.07
     customers
    -0.07
     现
    -0.07
     Rental
    -0.07
     Feed
    -0.07
     Cox
    -0.07
    greens
    -0.07
     backup
    -0.06
     wraps
    -0.06
    POSITIVE LOGITS
     rond
    0.07
    .rmtree
    0.06
    セン
    0.06
     sạn
    0.06
    ********************************************************************************
    0.06
     iliş
    0.06
    .WindowManager
    0.06
     책
    0.06
     rtrim
    0.06
     avere
    0.05
    Activations Density 0.058%

    No Known Activations