© 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. 26444
    Prev
    Next
    INDEX
    Explanations

    phrases related to feelings of being left behind or abandoned

    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
    515
    +0.09
    0.4%
    627
    +0.09
    0.4%
    605
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.09
    0.05
    566
    +0.09
    0.04
    102
    +0.09
    0.05
    Negative Logits
    oots
    -0.73
     Towards
    -0.68
    wcsstore
    -0.65
    issors
    -0.64
     Improvements
    -0.62
     Deaths
    -0.61
     Principles
    -0.61
     Advance
    -0.60
     Backup
    -0.60
     Excellence
    -0.59
    POSITIVE LOGITS
     scrambling
    1.20
     powerless
    1.19
     vulnerable
    1.18
     unable
    1.13
     stranded
    1.10
     paralyzed
    1.09
     baffled
    1.09
     susceptible
    1.07
     helpless
    1.05
     frustrated
    1.02
    Activations Density 0.438%

    No Known Activations