© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 29798
    Prev
    Next
    INDEX
    Explanations

    patterns of accusation and blame in various contexts

    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
    608
    +0.13
    0.6%
    393
    +0.10
    0.4%
    665
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.13
    0.04
    12
    +0.10
    0.03
    532
    +0.10
    0.03
    Negative Logits
    tn
    -0.74
    IJ
    -0.69
    cies
    -0.66
    tti
    -0.66
    hens
    -0.66
    mo
    -0.63
    glers
    -0.63
    hen
    -0.63
    thens
    -0.61
    goers
    -0.61
    POSITIVE LOGITS
     being
    0.96
     bias
    0.94
     racism
    0.90
     treason
    0.89
     heresy
    0.89
     witchcraft
    0.88
     involvement
    0.88
     having
    0.88
     complicity
    0.87
     hypocrisy
    0.87
    Activations Density 0.207%

    No Known Activations