© 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. 10-RES_POST_32K-OAI
    11. 31747
    Prev
    Next
    INDEX
    Explanations

    instances of the word "int" followed by numerical values

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

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    232
    +0.10
    0.5%
    369
    +0.10
    0.4%
    745
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.10
    0.02
    196
    +0.10
    0.01
    394
    +0.09
    0.01
    Negative Logits
     pseud
    -0.70
     Decay
    -0.68
     attachments
    -0.65
    awaru
    -0.62
     toes
    -0.62
     Mbps
    -0.61
     teenager
    -0.59
    Town
    -0.58
     footing
    -0.58
    ãģį
    -0.58
    POSITIVE LOGITS
    rep
    1.07
    roit
    1.02
    ros
    1.00
    ruct
    1.00
    ract
    0.96
    imate
    0.95
    ellig
    0.94
    ram
    0.94
    activ
    0.93
    rest
    0.91
    Activations Density 0.078%

    No Known Activations