© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 5562
    Prev
    Next
    INDEX
    Explanations

    references to loyalty and loyal followers or supporters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    242
    +0.10
    0.5%
    288
    +0.10
    0.5%
    685
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    242
    +0.10
    0.02
    288
    +0.10
    0.02
    247
    +0.09
    0.01
    Negative Logits
    phrine
    -0.76
    quick
    -0.67
    GAME
    -0.63
    OUT
    -0.63
    osed
    -0.62
    DERR
    -0.61
    paces
    -0.61
    oteric
    -0.60
    LOD
    -0.60
     concussion
    -0.59
    POSITIVE LOGITS
    ty
    1.01
     allegiance
    0.91
     oath
    0.90
    ties
    0.87
     loyalty
    0.84
    hips
    0.84
     servant
    0.82
    alty
    0.81
     loyal
    0.79
    iciary
    0.79
    Activations Density 0.074%

    No Known Activations