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. 74807
    Prev
    Next
    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash

    This neuron activates on verbs and terms used to describe a court or judge making or denying errors (e.g. “erred,” “denying”).

    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
     sy
    -0.07
     provinces
    -0.07
     الاخ
    -0.07
     Melbourne
    -0.06
     regions
    -0.06
    :test
    -0.06
     usado
    -0.06
     decode
    -0.06
     vidé
    -0.06
    decode
    -0.06
    POSITIVE LOGITS
    ement
    0.06
     Phật
    0.06
     مدت
    0.06
     iT
    0.06
    储
    0.06
    >{$
    0.06
    istic
    0.06
    (',')[
    0.06
     @(
    0.06
     who
    0.06
    Activations Density 0.005%

    No Known Activations