© 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. 4524
    Prev
    Next
    INDEX
    Explanations

    financial and numerical data related to subscriptions or donations

    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
    722
    +0.13
    0.6%
    755
    +0.11
    0.5%
    335
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.13
    0.03
    314
    +0.11
    -0.00
    699
    +0.09
    0.02
    Negative Logits
    orously
    -0.74
    allery
    -0.73
    imity
    -0.71
    enhagen
    -0.66
    Release
    -0.65
    plot
    -0.64
     Dise
    -0.62
    emale
    -0.62
    utsche
    -0.62
     bearer
    -0.62
    POSITIVE LOGITS
    soDeliveryDate
    0.81
    natureconservancy
    0.66
    hin
    0.64
     guiActive
    0.61
    Í
    0.56
     Armour
    0.55
     laundry
    0.54
    tons
    0.54
     wonders
    0.54
     scr
    0.54
    Activations Density 0.194%

    No Known Activations