© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 8217
    Prev
    Next
    INDEX
    Explanations

    comparative phrases related to oppositions or distinctions

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    592
    +0.14
    0.6%
    729
    +0.12
    0.5%
    186
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.14
    0.16
    274
    +0.12
    0.10
    253
    +0.09
    0.09
    Negative Logits
    hens
    -0.65
    personal
    -0.57
    mar
    -0.54
     Zombies
    -0.53
    bench
    -0.53
    bare
    -0.53
    hak
    -0.52
    claw
    -0.52
    ose
    -0.52
     Sample
    -0.52
    POSITIVE LOGITS
    .<
    0.89
    .</
    0.84
     latter
    0.82
     phenomenon
    0.80
    .[
    0.79
    Likewise
    0.73
     same
    0.73
    .
    0.71
    ,[
    0.71
    MSN
    0.70
    Activations Density 3.059%

    No Known Activations