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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 129598
    Prev
    Next
    INDEX
    Explanations

    Writing tips

    np_max-act · gemini-2.0-flash

    The neuron flags tokens in the assistant’s reply text immediately after the header marker—i.e. it activates on words in the model’s response rather than on the user’s input.

    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
    _refl
    -0.07
    тоф
    -0.07
     graphql
    -0.06
     Rigidbody
    -0.06
     Fahrenheit
    -0.06
    sembling
    -0.06
    SOFTWARE
    -0.06
     unserialize
    -0.06
    _Tis
    -0.06
    threshold
    -0.06
    POSITIVE LOGITS
     Sab
    0.07
     СРСР
    0.06
    Web
    0.06
    phil
    0.06
     second
    0.06
     />
    0.06
    جر
    0.06
    utions
    0.06
    -pills
    0.06
     añ
    0.06
    Activations Density 0.039%

    No Known Activations