© 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. 9082
    Prev
    Next
    INDEX
    Explanations

    occurrences of the abbreviation "RO" in various contexts

    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
    +0.13
    0.6%
    522
    +0.11
    0.5%
    409
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    67
    +0.13
    0.02
    622
    +0.11
    0.02
    326
    +0.10
    0.02
    Negative Logits
    manship
    -0.76
    ãĤº
    -0.70
    nai
    -0.69
     Norn
    -0.67
     Consent
    -0.63
    iveness
    -0.63
     Manson
    -0.63
     Crit
    -0.62
    foundland
    -0.61
    lihood
    -0.58
    POSITIVE LOGITS
    oster
    1.02
    zzi
    0.98
    ggy
    0.90
    ppo
    0.90
    aches
    0.88
    aching
    0.88
    elling
    0.85
    asted
    0.84
    issy
    0.83
    itzer
    0.82
    Activations Density 0.069%

    No Known Activations