© 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. 28909
    Prev
    Next
    INDEX
    Explanations

    legal terminology and references to documents related to laws and regulations

    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
    464
    +0.14
    0.7%
    102
    +0.12
    0.5%
    373
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.14
    0.07
    444
    +0.12
    0.05
    114
    +0.12
    0.06
    Negative Logits
     srfAttach
    -0.70
    uckland
    -0.64
    bryce
    -0.64
     Parables
    -0.64
    pedia
    -0.59
     Shift
    -0.57
    hoff
    -0.57
     Examiner
    -0.57
    allion
    -0.56
     Beware
    -0.56
    POSITIVE LOGITS
     which
    0.97
     referred
    0.89
     expressly
    0.85
     whatsoever
    0.85
     imported
    0.84
     arising
    0.84
     mentioned
    0.83
     occurring
    0.82
     incorporated
    0.81
     subscribed
    0.80
    Activations Density 0.585%

    No Known Activations