© 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. 32726
    Prev
    Next
    INDEX
    Explanations

    conditional statements regarding actions and outcomes

    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
    735
    +0.13
    0.6%
    229
    +0.12
    0.6%
    139
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.13
    0.12
    551
    +0.12
    0.10
    718
    +0.12
    0.09
    Negative Logits
     doubtless
    -0.72
     probably
    -0.72
     likely
    -0.67
    20439
    -0.66
     regret
    -0.62
    idia
    -0.61
    interstitial
    -0.61
    ãĤ¤ãĥĪ
    -0.58
     IDE
    -0.58
     cringe
    -0.58
    POSITIVE LOGITS
     careful
    0.70
     diligently
    0.69
     adequately
    0.68
     properly
    0.68
     sufficiently
    0.67
     harness
    0.65
    chie
    0.64
    ached
    0.63
    ealous
    0.63
     carefully
    0.63
    Activations Density 0.793%

    No Known Activations