© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 7029
    Prev
    Next
    INDEX
    Explanations

    structured data and performance metrics, particularly in reference to gaming or competitive contexts

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    373
    +0.18
    0.8%
    650
    +0.12
    0.5%
    308
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    153
    +0.18
    0.01
    622
    +0.12
    0.01
    573
    +0.12
    0.01
    Negative Logits
    nan
    -0.68
    ":[{"
    -0.68
    ocy
    -0.64
    vity
    -0.64
    rar
    -0.63
    uty
    -0.62
    :{
    -0.62
    atari
    -0.62
    00000
    -0.62
    jee
    -0.61
    POSITIVE LOGITS
     GOODMAN
    0.73
    ĻĤ
    0.73
    Ń·
    0.72
    ģ«
    0.70
     137
    0.66
    enhagen
    0.63
     Olymp
    0.61
     143
    0.60
     Mutant
    0.60
     Colossus
    0.60
    Activations Density 0.052%

    No Known Activations