© 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. 6317
    Prev
    Next
    INDEX
    Explanations

    numerical data and statistical analysis related to medical research

    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
    239
    +0.12
    0.5%
    439
    +0.11
    0.5%
    317
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    239
    +0.12
    0.21
    67
    +0.11
    0.21
    439
    +0.10
    0.13
    Negative Logits
     stuff
    -0.79
    ulhu
    -0.78
     perks
    -0.77
    estate
    -0.72
    illions
    -0.71
     Armageddon
    -0.70
    Joe
    -0.69
     Karma
    -0.67
    advertising
    -0.67
    TextColor
    -0.67
    POSITIVE LOGITS
     âĪ¼
    1.59
    âĢIJ
    1.28
     âī¥
    1.27
     phenotype
    1.25
     heterogeneity
    1.21
     âī¤
    1.19
     β
    1.17
     (âĪĴ
    1.16
     α
    1.13
     assay
    1.12
    Activations Density 15.429%

    No Known Activations