requirements.txt - jspace-nanogpt.git - J-space on nanoGPT: Jacobian Lens replication & critique. J-lens norms anti-correlate with token frequency (r ~ -0.61 to -0.69); frequency-matched synthetic pair + causal loss-reweighting tests.

summary refs log tree commit diff

path: root/requirements.txt

blob: d7bc5010d51444d70a9b6d6541aacd7c27c77a79 (plain)

# Runtime tested in the pytorch/pytorch:2.4.1-cuda11.8 image (see README).
torch==2.4.1
numpy
transformers==4.44.0   # only needed for the GPT-2-scale experiments (gpt2_jlens.py)