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

    Neuronpedia

    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. 44540
    Prev
    Next
    INDEX
    Explanations

    challenge

    np_max-act · gemini-2.0-flash

    This neuron does not respond to any tokens—it effectively never activates and so doesn’t detect any particular pattern.

    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
    anlık
    -0.07
    .ReLU
    -0.07
     glory
    -0.06
     BATCH
    -0.06
     eight
    -0.06
     güzel
    -0.06
    spor
    -0.06
    .onActivityResult
    -0.06
    opez
    -0.06
    aussian
    -0.06
    POSITIVE LOGITS
     challenged
    0.09
     undermine
    0.07
     reinforcing
    0.07
     Communities
    0.07
     challenging
    0.07
     argued
    0.07
     능
    0.07
    λή
    0.07
     undermining
    0.06
    刺
    0.06
    Activations Density 0.013%

    No Known Activations