© 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. 28239
    Prev
    Next
    INDEX
    Explanations

    technical terms and concepts related to programming and compilation, particularly involving C/C++ languages and libraries

    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
    326
    +0.13
    0.6%
    327
    +0.10
    0.5%
    141
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.13
    0.16
    255
    +0.10
    0.11
    435
    +0.10
    0.08
    Negative Logits
     Rollins
    -0.79
    iquid
    -0.78
    answered
    -0.77
     Weed
    -0.75
     saline
    -0.75
    ãĥ¯ãĥ³
    -0.74
     Narr
    -0.71
     Stamina
    -0.71
     Rowe
    -0.69
     Pelicans
    -0.69
    POSITIVE LOGITS
     gcc
    1.51
     GCC
    1.36
    LIB
    1.35
     libraries
    1.29
     compiler
    1.27
    lib
    1.23
     lib
    1.17
     compile
    1.14
    FLAG
    1.11
     compilation
    1.11
    Activations Density 1.912%

    No Known Activations