© 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. 26288
    Prev
    Next
    INDEX
    Explanations

    punctuated phrases or quotes within the text

    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
    373
    +0.10
    0.5%
    587
    +0.10
    0.5%
    123
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.10
    0.10
    160
    +0.10
    0.09
    107
    +0.10
    0.07
    Negative Logits
    Reports
    -0.86
    Company
    -0.83
    him
    -0.77
    someone
    -0.73
    President
    -0.71
    WHERE
    -0.71
    Parents
    -0.71
    Allah
    -0.71
    Cause
    -0.71
    ande
    -0.70
    POSITIVE LOGITS
    hots
    0.85
     bottleneck
    0.82
     flared
    0.80
    EStreamFrame
    0.79
     warp
    0.77
     snowball
    0.77
     pinch
    0.77
    mini
    0.76
    hidden
    0.76
     staggered
    0.75
    Activations Density 0.479%

    No Known Activations