© 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. 6930
    Prev
    Next
    INDEX
    Explanations

    numerical data related to financial figures and dates

    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
    314
    +0.20
    1.0%
    442
    +0.19
    0.9%
    160
    +0.17
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    442
    +0.20
    0.03
    627
    +0.19
    0.02
    107
    +0.17
    0.02
    Negative Logits
     FF
    -0.85
     Gear
    -0.83
     GT
    -0.82
     Arrow
    -0.81
     Sheep
    -0.81
     BB
    -0.81
     Play
    -0.80
     Rand
    -0.80
     Arm
    -0.80
     Store
    -0.79
    POSITIVE LOGITS
    2018
    1.57
    2017
    1.54
    2013
    1.53
    2014
    1.52
    2015
    1.48
    2016
    1.47
    2010
    1.47
    2012
    1.45
    2011
    1.45
    2008
    1.42
    Activations Density 0.047%

    No Known Activations