© 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. 28164
    Prev
    Next
    INDEX
    Explanations

    references to sources and additional information in academic writing

    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
    373
    +0.14
    0.6%
    628
    +0.12
    0.6%
    51
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.14
    -0.02
    211
    +0.12
    0.01
    67
    +0.11
    0.03
    Negative Logits
    henko
    -0.74
     Tactics
    -0.73
     Guilty
    -0.68
     Xiaomi
    -0.66
     Sands
    -0.65
     exec
    -0.64
    iency
    -0.64
     undrafted
    -0.64
     PvP
    -0.63
     redes
    -0.62
    POSITIVE LOGITS
     digit
    1.00
    ibrarian
    1.00
     Reprodu
    0.90
    ibli
    0.89
     reproduction
    0.87
     Dupl
    0.82
    Catalog
    0.81
     editions
    0.80
    DOC
    0.78
     reprodu
    0.78
    Activations Density 0.270%

    No Known Activations