© 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. 32697
    Prev
    Next
    INDEX
    Explanations

    key terms and phrases related to importance and opportunity

    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
    505
    +0.12
    0.5%
    17
    +0.11
    0.5%
    591
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    507
    +0.12
    0.12
    698
    +0.11
    0.10
    96
    +0.10
    0.09
    Negative Logits
     Philly
    -0.94
     Buffett
    -0.88
     Mines
    -0.87
     zinc
    -0.82
     Goldberg
    -0.80
     Berkshire
    -0.79
     Newtown
    -0.79
     Sioux
    -0.78
     Kand
    -0.78
     spinach
    -0.78
    POSITIVE LOGITS
     âĶĤ
    3.15
     âĶ
    1.86
     âĶľ
    1.86
     âĶľâĶĢâĶĢ
    1.84
    âĶ
    1.82
    âĶĢ
    1.38
     âĹı
    1.35
    âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ
    1.35
    âĶĢâĶĢ
    1.34
     âĹ
    1.28
    Activations Density 1.657%

    No Known Activations