© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 27493
    Prev
    Next
    INDEX
    Explanations

    variations of the word "know" and related concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    314
    +0.16
    0.7%
    420
    +0.12
    0.5%
    285
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    285
    +0.16
    0.02
    449
    +0.12
    0.02
    271
    +0.12
    0.02
    Negative Logits
     Mehran
    -0.79
     Canaver
    -0.71
     Statue
    -0.63
     Pax
    -0.63
     Jindal
    -0.62
     Pike
    -0.62
     cartoons
    -0.62
     pomp
    -0.62
     portraits
    -0.61
     Archdemon
    -0.59
    POSITIVE LOGITS
    ledged
    1.27
    ledge
    1.27
    lege
    1.27
    led
    0.99
    LED
    0.95
    how
    0.86
    held
    0.85
     ledge
    0.83
    leg
    0.83
    kn
    0.82
    Activations Density 0.222%

    No Known Activations