© 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. 26414
    Prev
    Next
    INDEX
    Explanations

    actions related to tasks or activities, especially those involving completion or assistance

    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
    652
    +0.13
    0.6%
    269
    +0.11
    0.5%
    245
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    351
    +0.13
    0.13
    665
    +0.11
    0.13
    271
    +0.10
    0.10
    Negative Logits
    geoning
    -0.85
    xtap
    -0.72
     unsur
    -0.68
    iosyncr
    -0.68
    icient
    -0.66
    athered
    -0.66
    erenn
    -0.66
    emonic
    -0.66
     governed
    -0.66
     strikingly
    -0.64
    POSITIVE LOGITS
     anyways
    1.05
     tonight
    1.04
     somet
    0.98
     anymore
    0.97
     my
    0.96
     tomorrow
    0.95
     something
    0.93
     anything
    0.92
    ?!"
    0.92
     :(
    0.91
    Activations Density 1.668%

    No Known Activations