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. 101256
    Prev
    Next
    INDEX
    Explanations

    figure references

    np_max-act · gemini-2.0-flash
    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
     worship
    -0.08
    aeper
    -0.06
    _Trans
    -0.06
    とい
    -0.06
     threads
    -0.06
     VID
    -0.06
    	product
    -0.06
    fusc
    -0.06
    sortable
    -0.06
    Command
    -0.06
    POSITIVE LOGITS
    ,<
    0.07
     erotische
    0.07
     шлях
    0.07
     Mal
    0.06
    】↵
    0.06
     witch
    0.06
    ド
    0.06
     ciphertext
    0.06
     Bali
    0.06
    oliberal
    0.06
    Activations Density 0.006%

    No Known Activations