© 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. 8-RES_POST_32K-OAI
    11. 8870
    Prev
    Next
    INDEX
    Explanations

    vertical bar symbols and associated numerical values

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    745
    +0.12
    0.5%
    373
    +0.12
    0.5%
    503
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    482
    +0.12
    0.01
    586
    +0.12
    0.01
    500
    +0.10
    0.01
    Negative Logits
    heed
    -0.73
    dfx
    -0.65
    ãĤ´ãĥ³
    -0.65
    ++++++++++++++++
    -0.61
    enegger
    -0.59
    woods
    -0.59
    uph
    -0.59
     Scorp
    -0.59
    eni
    -0.58
     Swordsman
    -0.57
    POSITIVE LOGITS
    biz
    0.63
    rition
    0.63
    idel
    0.61
     subscrib
    0.60
    ontent
    0.60
    otes
    0.59
     subscribe
    0.58
    iden
    0.58
    owler
    0.58
     scouting
    0.58
    Activations Density 0.084%

    No Known Activations