© 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. 26552
    Prev
    Next
    INDEX
    Explanations

    the conjunction "and" as well as phrases that emphasize connectivity and continuation in ideas

    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
    450
    +0.11
    0.5%
    264
    +0.10
    0.5%
    449
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    449
    +0.11
    0.04
    393
    +0.10
    0.04
    445
    +0.10
    0.03
    Negative Logits
    phot
    -0.73
     Ambro
    -0.73
    matically
    -0.73
    orsche
    -0.71
    ONSORED
    -0.69
    photo
    -0.68
    successfully
    -0.67
    agram
    -0.67
    aturdays
    -0.66
    ortmund
    -0.65
    POSITIVE LOGITS
     thou
    1.12
    areth
    0.99
     shalt
    0.98
     ye
    0.97
     righteousness
    0.96
     shall
    0.95
     everlasting
    0.94
     thy
    0.93
     sinners
    0.92
     Thou
    0.92
    Activations Density 0.198%

    No Known Activations