© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 26484
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "Justice" in various contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    167
    +0.10
    0.4%
    414
    +0.09
    0.4%
    132
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    132
    +0.10
    0.02
    224
    +0.09
    0.02
    325
    +0.09
    0.02
    Negative Logits
    cue
    -0.86
    terday
    -0.79
    gomery
    -0.74
     captcha
    -0.68
    isable
    -0.68
     Presbyter
    -0.67
    pload
    -0.65
    ecd
    -0.65
     paternity
    -0.63
     municip
    -0.62
    POSITIVE LOGITS
     Department
    1.01
     Dept
    0.85
     League
    0.82
     Anton
    0.76
    avia
    0.72
     Ministry
    0.70
    Ins
    0.69
     Gins
    0.64
     Blocks
    0.62
    Indust
    0.62
    Activations Density 0.216%

    No Known Activations