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. 129111
    Prev
    Next
    INDEX
    Explanations

    difficult

    np_max-act · gemini-2.0-flash

    technical descriptions of manufacturing processes for composite materials.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron spotlights language describing problems, difficulties or potential drawbacks (e.g. “problems,” “may arise,” “difficult,” “including”).

    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
    lı
    -0.07
    _help
    -0.07
    Encryption
    -0.07
    lopen
    -0.06
     try
    -0.06
    很多
    -0.06
     ayında
    -0.06
    (Create
    -0.06
     hồ
    -0.06
    >Login
    -0.06
    POSITIVE LOGITS
     ventil
    0.07
     ś
    0.06
    dney
    0.06
    PLICIT
    0.06
     ком
    0.06
     Fisher
    0.06
    uplicated
    0.06
     atas
    0.06
    (KeyCode
    0.06
    Rem
    0.06
    Activations Density 0.058%

    No Known Activations