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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 84402
    Prev
    Next
    INDEX
    Explanations

    /H

    np_max-act · gemini-2.0-flash

    The neuron detects email‐forward header routing tokens (e.g. the “/HOU/ECT” style tags in forwarded header lines).

    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
    .Ap
    -0.07
     Dire
    -0.06
     хлоп
    -0.06
     fraught
    -0.06
     unwitting
    -0.05
    ाहन
    -0.05
     Fulton
    -0.05
    ーレ
    -0.05
    べき
    -0.05
     paternal
    -0.05
    POSITIVE LOGITS
    ocol
    0.07
    stagram
    0.07
    ={"
    0.07
    Endpoint
    0.06
    terra
    0.06
     Circular
    0.06
     -$
    0.06
     Columbus
    0.06
    (mouse
    0.06
    	account
    0.06
    Activations Density 0.002%

    No Known Activations