© 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. 32659
    Prev
    Next
    INDEX
    Explanations

    negative numerical values or ratios in statistics

    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
    442
    +0.15
    0.7%
    83
    +0.09
    0.4%
    102
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    442
    +0.15
    0.04
    260
    +0.09
    0.03
    81
    +0.09
    0.03
    Negative Logits
     contingency
    -0.76
     laure
    -0.73
     Continental
    -0.69
     preference
    -0.69
     nodd
    -0.66
     vanity
    -0.65
     Shades
    -0.64
     Phi
    -0.64
     501
    -0.64
     candidates
    -0.63
    POSITIVE LOGITS
    yard
    1.04
    inning
    1.02
    game
    0.98
    straight
    0.97
    save
    0.93
    GAME
    0.88
    all
    0.87
    present
    0.87
    Game
    0.86
    yards
    0.86
    Activations Density 0.135%

    No Known Activations