© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 109383
    Prev
    Next
    INDEX
    Explanations

    various internet topics

    np_max-act · gemini-2.0-flash

    This neuron detects the user’s main topic or subject keywords (e.g. specific domains like “environmental politics,” “wind farms,” “food labeling,” “music,” “skin care,” “credit,” etc.).

    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
    -outs
    -0.08
    아파트
    -0.07
     alarming
    -0.07
     Mexico
    -0.07
    滅
    -0.06
     translation
    -0.06
    vec
    -0.06
     �
    -0.06
    歳
    -0.06
    dığını
    -0.06
    POSITIVE LOGITS
    .vendor
    0.07
     tow
    0.07
     melakukan
    0.06
     frac
    0.06
     addObject
    0.06
     konu
    0.06
    ़त
    0.06
    FirstOrDefault
    0.06
    .operations
    0.06
    .endswith
    0.06
    Activations Density 0.137%

    No Known Activations