© 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. 9121
    Prev
    Next
    INDEX
    Explanations

    references to specific entities and their attributes or achievements

    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
    678
    +0.11
    0.5%
    204
    +0.09
    0.4%
    29
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.11
    0.08
    588
    +0.09
    0.07
    482
    +0.09
    0.05
    Negative Logits
    wana
    -0.86
     Kinnikuman
    -0.80
    essee
    -0.78
    Ĥİ
    -0.75
    IUM
    -0.72
    eon
    -0.71
     Hampshire
    -0.70
     Norn
    -0.69
     COUNTY
    -0.69
    ãĤ¢ãĥ«
    -0.69
    POSITIVE LOGITS
    ig
    1.16
    igs
    1.02
    og
    0.98
    igraph
    0.97
    IG
    0.95
    LOG
    0.93
     LOG
    0.90
    ignt
    0.89
     Log
    0.86
    Log
    0.85
    Activations Density 3.251%

    No Known Activations