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

    Neuronpedia

    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. 6021
    Prev
    Next
    INDEX
    Explanations

    references to intelligence quotient (IQ) and its variations among different groups

    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
    14
    +0.10
    0.5%
    550
    +0.10
    0.4%
    755
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    375
    +0.10
    0.04
    415
    +0.10
    0.07
    439
    +0.10
    0.05
    Negative Logits
     »
    -0.79
     VIDE
    -0.71
    raits
    -0.68
     ...
    -0.68
     --
    -0.67
    ewitness
    -0.67
     ..."
    -0.67
     ours
    -0.63
     unforgettable
    -0.62
     our
    -0.62
    POSITIVE LOGITS
    ,[
    1.88
    ).[
    1.79
    .[
    1.73
    ".[
    1.64
    )[
    1.61
    :[
    1.57
    [
    1.56
    ][
    1.33
    "[
    1.31
    ."[
    1.27
    Activations Density 3.708%

    No Known Activations