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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 38462
    Prev
    Next
    INDEX
    Explanations

    ect, defects

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of the root “defect” (e.g. “defect,” “defects,” “defect-induced,” 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
     komunik
    -0.08
     railway
    -0.07
    �
    -0.07
     king
    -0.07
     knowledge
    -0.07
     network
    -0.07
     narration
    -0.07
     wire
    -0.07
     Waters
    -0.07
    	pw
    -0.07
    POSITIVE LOGITS
     defect
    0.12
     defective
    0.11
     defects
    0.11
    าก
    0.08
     Victor
    0.08
     In
    0.08
    _props
    0.07
    特
    0.07
    DC
    0.07
    -sc
    0.07
    Activations Density 0.005%

    No Known Activations