© 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. 8-RES_POST_32K-OAI
    11. 9120
    Prev
    Next
    INDEX
    Explanations

    specific numerical values or identifiers, particularly related to codes or statistics

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    89
    +0.11
    0.5%
    167
    +0.11
    0.5%
    406
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    220
    +0.11
    0.04
    104
    +0.11
    0.04
    260
    +0.10
    0.04
    Negative Logits
    veyard
    -0.76
     attendant
    -0.73
    creen
    -0.73
    urus
    -0.72
    otaur
    -0.71
    isson
    -0.69
    isoft
    -0.65
    ateurs
    -0.65
    cius
    -0.65
    ardless
    -0.65
    POSITIVE LOGITS
    00
    0.78
    625
    0.77
    OGR
    0.75
    dfx
    0.73
    acht
    0.71
    345
    0.70
    61
    0.69
    88
    0.69
    96
    0.68
    acl
    0.68
    Activations Density 0.215%

    No Known Activations