© 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 EvalsExportsinterp-engineNEWAPI 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. 26991
    Prev
    Next
    INDEX
    Explanations

    ideas about teaching and the importance of broader life lessons beyond specific subjects

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

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    74
    +0.14
    0.7%
    184
    +0.13
    0.6%
    40
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.14
    0.13
    431
    +0.13
    0.07
    246
    +0.11
    0.08
    Negative Logits
    uly
    -0.78
    ificantly
    -0.76
    ivalent
    -0.75
    ivably
    -0.71
    miah
    -0.70
    oaded
    -0.69
    haus
    -0.69
    oubted
    -0.68
    iHUD
    -0.66
    abase
    -0.64
    POSITIVE LOGITS
    everyone
    0.91
     ALWAYS
    0.91
    Everyone
    0.89
     EVERY
    0.89
     Everyone
    0.88
     teamwork
    0.86
     everyone
    0.85
     greatness
    0.83
    Everything
    0.83
     Everything
    0.83
    Activations Density 1.891%

    No Known Activations