© 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. 9010
    Prev
    Next
    INDEX
    Explanations

    negative numerical values or references to declines

    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
    373
    +0.22
    1.0%
    242
    +0.10
    0.5%
    482
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    537
    +0.22
    0.03
    427
    +0.10
    0.03
    639
    +0.09
    0.03
    Negative Logits
     pudding
    -0.70
     rhy
    -0.69
     downward
    -0.66
    swick
    -0.63
     dexter
    -0.60
     phys
    -0.60
     relegation
    -0.58
     Learns
    -0.58
     metast
    -0.58
     deviation
    -0.57
    POSITIVE LOGITS
    ECA
    0.81
    Asia
    0.76
    Secure
    0.75
    assies
    0.75
    111
    0.75
    CA
    0.73
    86
    0.73
    Pacific
    0.72
    Country
    0.72
    Americans
    0.71
    Activations Density 0.107%

    No Known Activations