© 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. 31755
    Prev
    Next
    INDEX
    Explanations

    technical specifications related to computer hardware

    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
    280
    +0.11
    0.5%
    550
    +0.10
    0.4%
    613
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.11
    0.14
    289
    +0.10
    -0.01
    197
    +0.10
    0.04
    Negative Logits
     polio
    -0.87
    externalActionCode
    -0.84
     dances
    -0.82
     spont
    -0.79
     Taxi
    -0.77
    natureconservancy
    -0.75
     Snapchat
    -0.73
     improvis
    -0.72
     coord
    -0.71
     Pasadena
    -0.71
    POSITIVE LOGITS
     ASUS
    1.62
     Asus
    1.56
     Corsair
    1.55
     MSI
    1.49
     motherboard
    1.43
     Acer
    1.29
     overcl
    1.27
     Lenovo
    1.21
     BIOS
    1.20
     PSU
    1.20
    Activations Density 2.851%

    No Known Activations