© 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. 30932
    Prev
    Next
    INDEX
    Explanations

    words or phrases that begin with "not" or "also"

    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
    +0.21
    1.0%
    176
    +0.14
    0.7%
    481
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    273
    +0.21
    0.08
    300
    +0.14
    0.07
    287
    +0.13
    0.06
    Negative Logits
    .","
    -0.81
     ..."
    -0.73
    ``
    -0.70
     thereto
    -0.67
    `
    -0.66
     wi
    -0.65
     accordingly
    -0.65
     ``(
    -0.65
    .''
    -0.64
     swe
    -0.62
    POSITIVE LOGITS
    odore
    1.13
    resa
    1.12
    mosp
    1.09
    endix
    0.98
    xiety
    0.97
     Conclusion
    0.96
    alyst
    0.94
    romeda
    0.94
    vantage
    0.92
    foundland
    0.91
    Activations Density 1.728%

    No Known Activations