© 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. 29127
    Prev
    Next
    INDEX
    Explanations

    contextual references to scientific or statistical concepts

    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
    +0.12
    0.6%
    502
    +0.10
    0.5%
    312
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.12
    0.18
    640
    +0.10
    0.15
    440
    +0.10
    0.12
    Negative Logits
    geries
    -0.79
    SPONSORED
    -0.75
    ĪĴ
    -0.75
    çīĪ
    -0.74
     smuggled
    -0.69
     unsub
    -0.68
     awaited
    -0.66
     rejo
    -0.64
     lockdown
    -0.63
     defect
    -0.63
    POSITIVE LOGITS
     Calcul
    1.08
     Higher
    1.02
     Average
    1.02
     averages
    0.96
    Higher
    0.96
     Percentage
    0.96
    Maximum
    0.95
     varies
    0.94
     Generally
    0.93
     Percent
    0.92
    Activations Density 2.536%

    No Known Activations