© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 38427
    Prev
    Next
    INDEX
    Explanations

    skill

    np_max-act · gemini-2.0-flash

    The neuron primarily detects the word “skill” (and its inflected forms) in the text.

    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
    oud
    -0.08
    átek
    -0.08
     ponder
    -0.07
     inhabited
    -0.07
     criticized
    -0.07
     东
    -0.07
     Feng
    -0.07
    ouz
    -0.06
    around
    -0.06
    ення
    -0.06
    POSITIVE LOGITS
     skills
    0.14
     skill
    0.13
     Skills
    0.12
     Skill
    0.11
    Skill
    0.11
     skilled
    0.10
    Skills
    0.09
    skill
    0.09
    .skills
    0.09
     habil
    0.08
    Activations Density 0.024%

    No Known Activations