Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 86344
    Prev
    Next
    INDEX
    Explanations

    Software build paths/errors

    np_max-act · gemini-2.0-flash

    This neuron detects filesystem path segments, i.e. tokens that are parts of file or directory paths (including slashes or backslashes).

    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
     Jac
    -0.07
     vulnerabilities
    -0.07
    \Factory
    -0.06
    Cab
    -0.06
     Mike
    -0.06
    ΙΚ
    -0.06
     tor
    -0.06
     veterinarian
    -0.06
     Tower
    -0.06
     lk
    -0.06
    POSITIVE LOGITS
    _Source
    0.07
     researching
    0.06
     FIG
    0.06
     Son
    0.06
     Liên
    0.06
    兩
    0.06
     Methods
    0.06
     structured
    0.06
    leshoot
    0.06
    .builder
    0.06
    Activations Density 0.006%

    No Known Activations