© 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. 7955
    Prev
    Next
    INDEX
    Explanations

    articles, particularly the word "a" and its frequency in the text

    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
    129
    +0.12
    0.6%
    635
    +0.10
    0.5%
    430
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.12
    0.12
    250
    +0.10
    0.05
    538
    +0.10
    0.04
    Negative Logits
     Dynamics
    -0.66
    idon
    -0.58
     deceased
    -0.56
     opponents
    -0.56
    lees
    -0.55
     notor
    -0.54
     trave
    -0.53
    UTC
    -0.53
    tube
    -0.52
     opponent
    -0.51
    POSITIVE LOGITS
    few
    1.15
     handful
    0.99
     fraction
    0.96
     marginally
    0.87
     few
    0.83
     cents
    0.82
     rudimentary
    0.80
     fleeting
    0.79
     occasional
    0.74
     trickle
    0.74
    Activations Density 0.295%

    No Known Activations