© 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. 26661
    Prev
    Next
    INDEX
    Explanations

    temporal markers, specifically dates and times

    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
    704
    +0.11
    0.5%
    718
    +0.09
    0.4%
    166
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    680
    +0.11
    0.03
    67
    +0.09
    0.03
    127
    +0.09
    0.02
    Negative Logits
    ansk
    -0.81
     WOM
    -0.72
    iets
    -0.72
     Gorge
    -0.69
     destro
    -0.69
    romeda
    -0.66
     behavi
    -0.66
    hide
    -0.65
     unden
    -0.64
     seiz
    -0.63
    POSITIVE LOGITS
     2020
    0.85
     1995
    0.83
    2000
    0.83
     2019
    0.83
     1996
    0.82
     2021
    0.82
     1997
    0.82
     2022
    0.82
     1994
    0.81
     1992
    0.80
    Activations Density 0.101%

    No Known Activations