© 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. 30469
    Prev
    Next
    INDEX
    Explanations

    concepts related to expansion and inclusion

    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
    547
    +0.13
    0.6%
    264
    +0.11
    0.5%
    171
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.13
    -0.00
    547
    +0.11
    0.05
    608
    +0.10
    0.12
    Negative Logits
    assium
    -0.73
    achev
    -0.67
    jac
    -0.66
     Reload
    -0.64
    glers
    -0.63
     partName
    -0.63
    Cla
    -0.62
     Oops
    -0.61
    henko
    -0.60
    heed
    -0.59
    POSITIVE LOGITS
     encompass
    1.33
     beyond
    1.19
     broader
    1.06
     wider
    1.06
     encompasses
    1.00
     encomp
    0.98
     borders
    0.97
     extends
    0.96
     realms
    0.95
     suburbs
    0.92
    Activations Density 3.544%

    No Known Activations