11 Commits

Author SHA1 Message Date
983286e216 📚 batch norm 2021-02-01 14:43:11 +05:30
20d2e27a3c 📚 ffn notes 2021-01-25 22:09:11 +05:30
2927aa217b gpt notes 2021-01-14 09:37:12 +05:30
4c769128cb autoregression trainer notes 2021-01-14 08:46:41 +05:30
774a72e0bf gpt 2021-01-13 12:54:43 +05:30
ce190701e8 global step 2021-01-07 21:11:50 +05:30
c124348b14 gradient clipping 2021-01-07 13:33:32 +05:30
dfefde657b no loss smoothing 2021-01-07 12:02:19 +05:30
b9da12ee3f d_model 2021-01-06 17:24:09 +05:30
4fe8392a8d auto regression common exp 2020-12-27 07:30:20 +05:30
c64b98e390 basic auto regression experiment for reuse 2020-12-26 21:12:44 +05:30