1
0
mirror of https://github.com/karpathy/minGPT synced 2024-11-15 19:10:39 +01:00

Commit Graph

  • 8405765843 bug fix Ziliang Peng 2024-04-27 22:26:45 -0700
  • c447ee0e9f fix the no_grad bias issue Ziliang Peng 2024-04-27 22:10:45 -0700
  • db3301a1e1 tidy import Ziliang Peng 2024-04-27 21:55:24 -0700
  • 2a22fff05e super small cleanup Ziliang Peng 2024-04-27 18:38:41 -0700
  • 9568f28e3e cancel clip_grad_norm_ Ziliang Peng 2024-04-27 18:31:35 -0700
  • 230ceaea8c add simple NOTE Ziliang Peng 2024-04-27 18:26:16 -0700
  • 2b1f9ba8f6 some cleanup of adder Ziliang Peng 2024-04-27 18:19:03 -0700
  • b49cbd9e0d some cleanup Ziliang Peng 2024-04-27 18:13:39 -0700
  • 6ee157ec24 cancel init_w Ziliang Peng 2024-04-27 18:08:19 -0700
  • 42838742bf cancel decay Ziliang Peng 2024-04-27 18:03:54 -0700
  • 63ee70890a fixed the tinygrad no-grad issue. moving bias as local var Ziliang Peng 2024-04-27 17:47:54 -0700
  • a1bf2b4b85 print total params count Ziliang Peng 2024-04-27 17:28:24 -0700
  • 2feca7f751 bring back the dropout Ziliang Peng 2024-04-27 17:13:58 -0700
  • c24d17a2b2 a little bit of cleanup Ziliang Peng 2024-04-27 16:46:12 -0700
  • 29bdc7d9d9 a bersion that almost runs except a few tensor without grad Ziliang Peng 2024-04-27 16:22:25 -0700
  • c4abaae11e wip Ziliang Peng 2024-04-27 11:28:30 -0700
  • 885d3ac488 some more stupid wip Ziliang Peng 2024-04-26 20:41:47 -0700
  • 4698a9fbfe wip Ziliang Peng 2024-04-26 18:29:32 -0700
  • 114f2b621f
    Merge pull request #2 from ziliangpeng/v--dataloader Victor Ziliang Peng 2024-04-26 20:12:07 -0700
  • eb3aba320c use proxy DataLoader Ziliang Peng 2024-04-26 17:43:56 -0700
  • 5638752b5b
    Merge pull request #1 from ziliangpeng/v--rename Victor Ziliang Peng 2024-04-26 16:22:36 -0700
  • b47cf06d1c git ignore some files Ziliang Peng 2024-04-26 16:18:13 -0700
  • 149c1b2efb rename a bunch of code ref Ziliang Peng 2024-04-26 15:59:47 -0700
  • bbd21207d8 rename main code folder Ziliang Peng 2024-04-26 15:56:46 -0700
  • 44e3696cbf
    Merge 34d9559966d28730389a5e2424f9306ee03ccb2d into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Westen-M 2023-11-29 01:09:41 +0100
  • 34d9559966 UL Attempts --- 2023-11-28 16:27:27 -0700
  • 8597f458d9 UL Basics with denoiser --- 2023-11-27 18:14:01 -0700
  • cb77f8f334 Gitignore and pile file --- 2023-11-16 16:32:37 -0700
  • ec38d4fbbf Working state --- 2023-11-16 16:30:52 -0700
  • 26b50034a6
    Merge 88ba2a4bf593988e36d79eb8dc670bc2a403102b into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Amnon Bleich 2023-08-21 11:44:43 +0200
  • 88ba2a4bf5 bug fix - remove attn.bias keys from GPT state dict in 'from_pretrined'. otherwise assertion fails Amnon Bleich 2023-08-21 11:39:12 +0200
  • 92ea014b4d
    Merge f37020777cbdd5a0072f8eabd859131d05f0ce9f into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Joseph Catrambone 2023-08-05 21:06:51 -0700
  • f37020777c Check off a todo in utils: add a method freeze() which returns a frozen config. JosephCatrambone 2023-08-05 20:44:07 -0700
  • c39af9393c
    Merge 8e2cc15edc0bf33111087aa271113aef87f3dc57 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 prasad83 2023-07-01 12:40:20 -0700
  • 8e2cc15edc
    Fixed configuration load for generate_repl prasad83 2023-06-30 13:39:12 +0530
  • 7e3fb8f931
    Added generator repl for using adder model. prasad83 2023-06-30 13:26:05 +0530
  • 159b7cd701
    Merge a5accfa42552bc683144062f08965d230ea7cff6 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 rjarun8 2023-06-28 21:56:00 +0530
  • a5accfa425 Rename transformer layers variable rjarun8 2023-06-28 21:51:29 +0530
  • cd6e96d168 Rename transformer layers variable rjarun8 2023-06-28 21:40:21 +0530
  • b3c609956b
    Merge 0386f75899f7268f4a414dea4a92716421e1c267 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Dayne 2023-02-26 21:59:20 -0800
  • 0386f75899
    Update bpe.py Dayne 2023-02-26 21:57:56 -0800
  • 9b72a303de
    Merge 1bcf2eca101347d7bcc4fc52499d7b460858da87 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 kukuquack 2023-02-17 15:13:37 +0200
  • 1bcf2eca10
    Adding a requirements.txt file kukuquack 2023-02-17 15:11:48 +0200
  • f26e1b6c8f
    Update README.md hoangkimthuc 2023-02-05 21:59:41 +0700
  • ac863c0a60
    Merge 756c302e804d4a680b16831170a917a82434469b into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Clive Chan 2023-01-17 18:52:23 -0800
  • 756c302e80
    Zero-grad more aggressively to save memory Clive Chan 2023-01-17 18:50:04 -0800
  • 0b6982b6c2
    Merge b45b695029ef41bafd333b084f8e74122c25ff2a into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Daniel Gross 2023-01-17 13:48:21 +0400
  • ab921c597d
    Merge 59047a6cabfbb2d2a5b808bd4606162efda14d97 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Dmitry Nikolayev 2023-01-15 17:53:40 -0500
  • 9dbda0ee5c
    Merge a362aa626be3926fe198aabd0a2847da0407bb83 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Gavia Gray 2023-01-12 14:37:53 -0500
  • 42919c20ce
    Merge adf1e57252eee4b0d139060f8953e74803a83cc4 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Younes Belkada 2023-01-11 22:14:02 +0800
  • 6e3906ad09
    Merge 7c22554ef65980147419e8a9393a228cf633db8a into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Equim 2023-01-11 22:13:30 +0800
  • 2d1cf09276
    Merge 62b6978a7ba7829fd9c0e67f01404cdb8a576489 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Costa Huang 2023-01-10 11:17:41 -0500
  • 62b6978a7b minor refactor Costa Huang 2023-01-10 11:14:05 -0500
  • 4ce48ae603 Minor refactor on variable names Costa Huang 2023-01-10 11:08:28 -0500
  • c5ac138f0d
    Merge 80237613f2826a0a67333d7490afdb133d1ca8ff into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Mohamed Rashad 2023-01-09 08:06:25 +0200
  • 80237613f2 Better README.md Mohamed Rashad 2023-01-09 06:05:40 +0000
  • 68ab54af3b Better README.md Mohamed Rashad 2023-01-09 06:05:27 +0000
  • 5fbe921f13
    Merge 325000e6314dd1a4fa20bf076e873b81b0aa8e28 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Ikko Eltociear Ashimine 2023-01-08 22:50:54 +0200
  • 97ad95f914
    Merge 26d5e00274e1466a31fc946d4ab0284b1cdb4330 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Benjamin Schulz 2023-01-08 22:50:44 +0200
  • 16ccb8d248
    Merge b59643b884ffae5f891f53ce779c6db1dea372a4 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Dmitry Nikolayev 2023-01-09 00:53:02 +0800
  • 9766354d35
    Merge 4759df4fba323caf5a17c634ed5a0460db775820 into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Bora Gökbakan 2023-01-09 00:53:01 +0800
  • edaf232041
    Merge 06b7c3200efe6e92d4f31d2b2a376401207a282e into 37baab71b9abea1b76ab957409a1cc2fbfba8a26 Chinhai Hour 2023-01-09 00:53:01 +0800
  • 37baab71b9
    Update README.md master Andrej 2023-01-08 08:50:20 -0800
  • 4759df4fba unittests boragokbakan 2023-01-07 22:12:32 +0100