Commit Graph

  • a31a3b2e92 + MHA_shift BlinkDL 2021-08-14 03:01:05 +0800
  • 4096fff9ee
    Update README.md PENG Bo 2021-08-14 00:25:41 +0800
  • 12ba06216d
    Update README.md PENG Bo 2021-08-14 00:24:37 +0800
  • 639de69256
    Create CITATION.cff PENG Bo 2021-08-14 00:22:14 +0800
  • 994170685b
    Update README.md PENG Bo 2021-08-14 00:06:25 +0800
  • 3b9005ea11 RWKV: now faster and less params 0.01 BlinkDL 2021-08-13 18:39:24 +0800
  • 546114c6a5 still use layernorm for everything BlinkDL 2021-08-13 15:55:57 +0800
  • c68ea168b1
    Update README.md PENG Bo 2021-08-13 14:21:50 +0800
  • 73a63e175f
    Update README.md PENG Bo 2021-08-13 13:56:30 +0800
  • 2df321d3f4
    Update README.md PENG Bo 2021-08-13 13:56:06 +0800
  • 6e2ba61d95
    Update README.md PENG Bo 2021-08-13 13:54:24 +0800
  • cd9b352b45
    Update README.md PENG Bo 2021-08-13 11:48:22 +0800
  • d2b100c2ac
    Update README.md PENG Bo 2021-08-13 11:46:31 +0800
  • 8af6289d0c
    Update README.md PENG Bo 2021-08-13 03:07:13 +0800
  • fd098b1d2e small update BlinkDL 2021-08-13 02:25:43 +0800
  • 3b01c8c3cf
    Update README.md PENG Bo 2021-08-12 21:17:10 +0800
  • 65eda0f915 no message BlinkDL 2021-08-12 21:14:17 +0800
  • 3b60c5b266 add wandb, and rename variables BlinkDL 2021-08-12 20:56:31 +0800
  • 440bebff1a fixed nan in large models BlinkDL 2021-08-12 12:15:27 +0800
  • f80ff53595
    Update README.md PENG Bo 2021-08-12 12:00:42 +0800
  • 62e2cb06d6 fixing nan in large models BlinkDL 2021-08-11 22:11:12 +0800
  • d699a69169 misc improvements BlinkDL 2021-08-11 19:00:02 +0800
  • 6266f481da minor changes BlinkDL 2021-08-11 15:53:44 +0800
  • 88297e7949
    Update README.md PENG Bo 2021-08-11 15:32:35 +0800
  • 89eab46e60 + info BlinkDL 2021-08-11 14:54:51 +0800
  • e9fbd9bf70 remove layernorm -> better RWKV BlinkDL 2021-08-11 14:39:57 +0800
  • 55405c57d0 better splitting of words BlinkDL 2021-08-10 12:53:21 +0800
  • 01d6972f4f now works for word-level LM BlinkDL 2021-08-09 22:01:44 +0800
  • 64fdb61056
    Update README.md PENG Bo 2021-08-09 19:38:45 +0800
  • 959115a7e6
    Update README.md PENG Bo 2021-08-09 19:25:24 +0800
  • 447eae5841 add MHA-plus model BlinkDL 2021-08-09 16:51:01 +0800
  • bcd4adb781
    Update README.md PENG Bo 2021-08-09 15:52:13 +0800
  • 1035a7438e
    Update README.md PENG Bo 2021-08-09 14:40:08 +0800
  • 4c6db5607c
    Update README.md PENG Bo 2021-08-09 14:31:02 +0800
  • aa4e2a68f4 first commit BlinkDL 2021-08-09 13:52:19 +0800
  • d21af78c97
    Initial commit PENG Bo 2021-08-08 14:05:27 +0800