© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 32715
    Prev
    Next
    INDEX
    Explanations

    phrases and words related to regulatory and policy discussions

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    362
    +0.14
    0.6%
    608
    +0.12
    0.6%
    408
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.14
    0.06
    669
    +0.12
    0.05
    362
    +0.12
    0.05
    Negative Logits
    à¼
    -1.14
    adj
    -0.82
    姫
    -0.81
    lite
    -0.80
    lime
    -0.79
    û
    -0.78
    inen
    -0.74
    retty
    -0.74
    NES
    -0.74
    linger
    -0.73
    POSITIVE LOGITS
     matters
    0.91
     whether
    0.87
     finances
    0.86
     refunds
    0.86
     reliability
    0.85
     topics
    0.84
     legality
    0.83
     accuracy
    0.83
     accessibility
    0.82
     acquisitions
    0.81
    Activations Density 0.325%

    No Known Activations