© 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. 30305
    Prev
    Next
    INDEX
    Explanations

    terms related to legal rulings and concepts within the justice system

    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
    103
    +0.13
    0.6%
    516
    +0.12
    0.5%
    373
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    103
    +0.13
    0.09
    756
    +0.12
    0.01
    439
    +0.11
    0.09
    Negative Logits
    Legend
    -0.73
    ï¸ı
    -0.72
    orge
    -0.72
     Reborn
    -0.70
     TEAM
    -0.69
     Finish
    -0.68
     Dwarf
    -0.67
     Patriarch
    -0.67
     Blueprint
    -0.66
     Reboot
    -0.66
    POSITIVE LOGITS
     appellant
    1.23
     plaintiffs
    1.18
     plaintiff
    1.16
     statute
    1.16
     §
    1.15
     petitioner
    1.15
     constitutionally
    1.14
     defendant
    1.12
     Plaint
    1.12
     appell
    1.11
    Activations Density 3.835%

    No Known Activations