© 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. 28353
    Prev
    Next
    INDEX
    Explanations

    phrases related to judgment and accountability

    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
    116
    +0.13
    0.6%
    497
    +0.11
    0.5%
    454
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.13
    0.13
    55
    +0.11
    0.11
    566
    +0.11
    0.09
    Negative Logits
     convergence
    -0.84
     divergence
    -0.80
    PsyNetMessage
    -0.80
    isSpecialOrderable
    -0.69
     reproduction
    -0.66
     specificity
    -0.65
     sightings
    -0.65
     Improvement
    -0.64
     encounters
    -0.64
     Languages
    -0.64
    POSITIVE LOGITS
     jailed
    1.50
     punished
    1.50
     prosecuted
    1.40
     fined
    1.38
     sentenced
    1.34
     acquitted
    1.29
     imprisoned
    1.28
     convicted
    1.26
     expelled
    1.24
     reprim
    1.24
    Activations Density 0.751%

    No Known Activations