© 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. 10-RES_POST_32K-OAI
    11. 30418
    Prev
    Next
    INDEX
    Explanations

    commands related to enabling or disabling features and functions

    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
    59
    +0.13
    0.6%
    288
    +0.11
    0.5%
    475
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    59
    +0.13
    0.02
    326
    +0.11
    0.03
    507
    +0.10
    0.02
    Negative Logits
    apest
    -0.89
    sworth
    -0.77
    axies
    -0.76
    nant
    -0.76
    itual
    -0.72
     Fernand
    -0.72
    ioxide
    -0.71
    geon
    -0.69
    llah
    -0.68
    arium
    -0.67
    POSITIVE LOGITS
    ment
    0.78
     javascript
    0.77
     automatic
    0.73
    aucus
    0.66
    Disable
    0.65
    full
    0.64
     sprink
    0.64
     Automatic
    0.64
     anonymity
    0.64
    Scroll
    0.62
    Activations Density 0.161%

    No Known Activations