© 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 EvalsExportsAPI 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. 6715
    Prev
    Next
    INDEX
    Explanations

    sections related to editing or revisions in content

    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
    432
    +0.11
    0.5%
    373
    +0.10
    0.5%
    645
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    555
    +0.11
    0.05
    702
    +0.10
    0.05
    373
    +0.10
    -0.01
    Negative Logits
    ktop
    -0.69
     manoeuv
    -0.61
    ullivan
    -0.61
    rake
    -0.61
    chy
    -0.61
    hower
    -0.59
     lane
    -0.59
     pent
    -0.58
    kefeller
    -0.57
    gart
    -0.57
    POSITIVE LOGITS
    <|endoftext|>
    0.76
    Appearances
    0.74
    Ĥİ
    0.74
    pmwiki
    0.73
     Tribunal
    0.71
    Below
    0.67
     Standing
    0.66
     Award
    0.62
    Notable
    0.62
     Notably
    0.61
    Activations Density 0.076%

    No Known Activations