© 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. 10-RES_POST_32K-OAI
    11. 28255
    Prev
    Next
    INDEX
    Explanations

    numerical data and relationships in a mathematical context

    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
    314
    +0.22
    1.1%
    442
    +0.18
    0.9%
    160
    +0.18
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.22
    0.07
    107
    +0.18
    0.06
    442
    +0.18
    0.06
    Negative Logits
     roundup
    -0.92
     metic
    -0.87
     thrill
    -0.83
     spir
    -0.81
     ende
    -0.81
     quir
    -0.80
     nodd
    -0.80
     volunt
    -0.78
     poaching
    -0.78
     accomp
    -0.78
    POSITIVE LOGITS
    etc
    1.60
    Tan
    1.35
    Various
    1.29
    Salt
    1.27
    Rain
    1.26
    Basic
    1.26
    MIC
    1.25
    Pa
    1.24
    Jar
    1.24
    Fred
    1.23
    Activations Density 0.310%

    No Known Activations