Step 23774: loss=0.1861

Step 23149: loss=0.1864

Step 22524: loss=0.1868

Training step 22000

Step 21899: loss=0.1871

Step 21274: loss=0.1875

Step 20649: loss=0.1879

Step 20024: loss=0.1883

Training step 20000

Step 19399: loss=0.1888

Step 18774: loss=0.1893

Step 18149: loss=0.1898

Training step 18000

Step 17524: loss=0.1904

Step 16899: loss=0.1911

Step 16274: loss=0.1918

Training step 16000

Step 15649: loss=0.1926

Step 15024: loss=0.1935

Step 14399: loss=0.1945

Training step 14000

Step 13774: loss=0.1956

Step 13149: loss=0.1968

Step 12524: loss=0.1982

Training step 12000

Step 11899: loss=0.1997

Step 11274: loss=0.2015

Step 10649: loss=0.2034

Step 10024: loss=0.2056

Training step 10000

Training step 8774

Step 8750: loss=0.2074

Step 8125: loss=0.2097

Training step 8000

Step 7500: loss=0.2120

Step 6875: loss=0.2130

Step 6250: loss=0.2141

Training step 6000

Step 5625: loss=0.2159

Rename checkpoint_step_5000.pt to model.pt