© 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. 27579
    Prev
    Next
    INDEX
    Explanations

    references to server configurations and performance metrics

    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
    378
    +0.13
    0.6%
    245
    +0.12
    0.5%
    38
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.13
    0.14
    536
    +0.12
    0.04
    347
    +0.11
    0.04
    Negative Logits
     Tak
    -0.83
     hemp
    -0.82
     cannabinoid
    -0.82
    eco
    -0.82
     artic
    -0.80
     Suzuki
    -0.79
    Toy
    -0.78
     Elliott
    -0.75
     Autob
    -0.74
    arijuana
    -0.74
    POSITIVE LOGITS
     Server
    2.10
    server
    2.05
     server
    2.05
    Server
    1.98
     servers
    1.87
     Ser
    1.76
    erver
    1.66
    Ser
    1.50
    ser
    1.48
     Client
    1.33
    Activations Density 3.072%

    No Known Activations