Commit Graph

Select branches

Hide Pull Requests

master

#103

#105

#106

#108

#109

#110

#111

#112

#113

#115

#116

#119

#121

#122

#123

#126

#127

#128

#128

#129

#13

#130

#131

#132

#133

#135

#137

#14

#140

#141

#141

#142

#144

#145

#146

#147

#147

#149

#15

#15

#151

#151

#153

#154

#155

#156

#159

#159

#160

#161

#161

#162

#165

#169

#17

#172

#172

#173

#174

#175

#179

#18

#180

#181

#182

#184

#185

#186

#19

#190

#191

#192

#194

#195

#197

#2

#201

#201

#204

#204

#205

#21

#217

#224

#226

#227

#228

#23

#230

#232

#233

#234

#235

#236

#237

#238

#239

#24

#240

#241

#242

#243

#244

#246

#247

#249

#251

#252

#252

#253

#253

#255

#256

#256

#258

#258

#259

#261

#262

#263

#265

#267

#268

#269

#27

#270

#271

#274

#275

#275

#276

#279

#280

#281

#282

#286

#29

#290

#294

#294

#295

#296

#296

#298

#299

#3

#3

#30

#301

#301

#306

#306

#307

#307

#308

#308

#309

#309

#31

#31

#310

#310

#311

#311

#312

#312

#316

#316

#317

#317

#318

#318

#319

#319

#32

#32

#322

#322

#323

#324

#324

#325

#325

#326

#326

#327

#327

#328

#328

#329

#33

#330

#333

#333

#336

#34

#342

#342

#345

#345

#346

#348

#348

#35

#350

#351

#351

#352

#353

#355

#356

#357

#357

#36

#38

#38

#39

#4

#40

#40

#41

#43

#46

#48

#49

#5

#50

#51

#52

#53

#53

#54

#55

#55

#56

#59

#59

#6

#61

#62

#63

#63

#64

#66

#67

#70

#74

#75

#80

#81

#84

#85

#85

#86

#88

#89

#9

#90

#90

#91

#93

#93

#94

#95

#97

#98

#99

Mono Color

• e3393045a4
  Merge 337d9649c6 into 4a87a0d19f Dipesh Babu 2025-12-05 17:38:03 -0600
• 337d9649c6
  Merge branch 'karpathy:master' into master Dipesh Babu 2025-12-05 18:20:35 -0500
• 2288750906
  Merge 59ed9392ed into 4a87a0d19f Eyal Frishman (Nvidia) 2025-12-05 20:11:59 +0200
• 59ed9392ed Add pre-commit documentation to README and GitHub workflow Eyal Frishman 2025-11-25 17:27:54 +0200
• 449494c8b6 Fix (automatically) all pre-commit errors Eyal Frishman 2025-12-05 18:33:00 +0200
• 6587063479 Add pre-commit hooks for code formatting (not yet executed) Eyal Frishman 2025-12-05 18:26:55 +0200
• da26e7408e
  Merge 8b75ee2d4a into 4a87a0d19f Vlad Pavlov 2025-12-04 23:32:02 -0600
• 8b75ee2d4a Invitation to Mr. Karpathy to speak in front of 180+ engineering leaders on Dec 16th in Menlo Park about reimagining SDLC with Copilots and Vibe-Coding. Vlad Pavlov 2025-12-04 23:28:35 -0600
• 0d4f0250df
  标注：speedrun.sh - 核心逻辑 / 设计思路 / 代码规范 Zoey-u 2025-12-04 17:29:07 +0800
• 4528ecc97f benchmark for optimisations diana-bi 2025-12-03 21:48:01 +0330
• 96ec37e5fd Fix epsilon scaling in DistAdamW to match standard AdamW Charles Weill 2025-12-02 12:26:41 -0800
• 55f8d4acf2 Fix OOM in Japanese tokenizer training by reducing max_chars karaage0703 2025-12-02 12:09:24 +0900
• 1edc13ddb2
  Merge a8847a0f83 into 4a87a0d19f KimYeongHyeon 2025-12-02 10:48:53 +0900
• a8847a0f83
  Fix script comment to reference correct file KimYeongHyeon 2025-12-02 10:46:20 +0900
• 2adcc95c4e
  Merge branch 'master' into refactor-vertex-ai-pipelines javasoup 2025-12-01 20:07:43 -0500
• 13001597c2
  Success on Vertex Pipelines Nuno Pereira 2025-12-01 19:59:58 -0500
• 86a6cf6668 Add Kiro steering documents for project context karaage0703 2025-12-01 21:35:09 +0900
• e1e836763e Add Japanese language support for nanochat karaage0703 2025-12-01 21:26:34 +0900
• 1b247bff81
  Merge e50896dcdc into 4a87a0d19f Paweł Krefta 2025-12-01 01:21:02 +0100
• e50896dcdc Fix conversation scroll to bottom on some browsers + remove duplicated padding Pawel Krefta 2025-12-01 01:12:09 +0100
• 62cfe4d4c3 Instrument main gpt model with logging Wollaston 2025-11-30 14:11:33 -0500
• 8a40915246 Create logging wrapper for function name, args, and kwargs Wollaston 2025-11-30 14:05:48 -0500
• e85c309235 Add loguru Wollaston 2025-11-30 14:05:31 -0500
• 606314b05e
  Merge 06677c30e0 into 4a87a0d19f deepbuilder 2025-11-28 15:22:42 -0500
• 06677c30e0
  Refactor dimension validation for KV cache deepbuilder 2025-11-28 15:22:18 -0500
• a770dcef2e
  Fix kv_cache indexing to explicitly include head dimension deepbuilder 2025-11-28 15:00:14 -0500
• 3fc31e56ba
  Merge 24c04f0ca7 into 4a87a0d19f kiankyars 2025-11-26 19:00:55 -0700
• 24c04f0ca7 fix merge conflict manually, cursor fails Kian Kyars 2025-11-26 19:00:38 -0700
• d745be732e Support append mode in Report.log Cursor Agent 2025-11-27 01:41:53 +0000
• b113978aae
  Merge b06bedac08 into 4a87a0d19f marked23 2025-11-26 18:10:51 +0100
• b06bedac08
  remove empty line Sofie Van Landeghem 2025-11-26 18:09:23 +0100
• 2c6a007e3c
  Fix race condition in save_checkpoint for non-zero ranks zzF 2025-11-26 14:05:11 +0800
• 1eaaba1c64 Add ToDo.md for tasks and roadmap google-labs-jules[bot] 2025-11-24 19:10:55 +0000
• 74b03694b1
  Merge pull request #26 from LokiMetaSmith/tinyrun-integration Lawrence R Kincheloe III 2025-11-24 13:00:12 -0600
• e6ce7a06b0
  Merge branch 'master' into master Fred Bliss 2025-11-24 07:00:46 -0600
• 455c3f070b
  Merge pull request #1 from fblissjr/add-comprehensive-documentation Fred Bliss 2025-11-24 06:58:07 -0600
• 51927a9e60 feat: Add comprehensive end-to-end documentation google-labs-jules[bot] 2025-11-24 12:57:49 +0000
• 6838699bd4 Update tinyrun.sh with AMD Strix Halo optimizations google-labs-jules[bot] 2025-11-23 22:08:02 +0000
• 146ce1a8f6 Add tinyrun.sh script for single GPU/CPU training google-labs-jules[bot] 2025-11-23 21:53:38 +0000
• 490d517d89
  Merge 02b22a5a13 into 4a87a0d19f Anton Chechetka 2025-11-23 17:57:17 +0100
• 02b22a5a13 Fix relative difference sign in scripts/tok_eval.py Anton Chechetka 2025-11-23 17:51:18 +0100
• cbdca27e27
  Merge pull request #24 from LokiMetaSmith/fix-amd-triton-reinstall Lawrence R Kincheloe III 2025-11-23 10:03:31 -0600
• bbc816dc77 Reduce base_train batch size and set PYTORCH_HIP_ALLOC_CONF google-labs-jules[bot] 2025-11-23 16:03:02 +0000
• 52c7d23a63
  Merge 2f4f20862d into 4a87a0d19f kiankyars 2025-11-23 08:11:54 -0700
• 2f4f20862d add back comment Kian Kyars 2025-11-23 08:09:28 -0700
• 1d719a7c94 add back hugging face tokenizer Kian Kyars 2025-11-23 08:07:52 -0700
• d28d69f3ea reduce list redundancy Kian Kyars 2025-11-23 08:07:11 -0700
• 6398858ea9
  Merge 16788eed3c into 4a87a0d19f Sitananda Prasad 2025-11-23 20:14:26 +0530
• 16788eed3c fix(model): apply float32 cast before logits softcapping spjosyula 2025-11-23 20:12:09 +0530
• bfa37c8723
  Merge 3a611f0821 into 4a87a0d19f Anton Chechetka 2025-11-23 12:59:58 +0100
• 3a611f0821 Clean up copypaste in GPT.forward() Anton Chechetka 2025-11-23 12:51:36 +0100
• e14d7ba6bf
  Merge pull request #23 from LokiMetaSmith/fix-amd-triton-reinstall Lawrence R Kincheloe III 2025-11-23 02:28:00 -0600
• 41ba458c3b Explicitly enable allow_tf32 in nanochat/common.py google-labs-jules[bot] 2025-11-23 08:27:13 +0000
• 40ef6e81a9
  Merge pull request #22 from LokiMetaSmith/fix-amd-triton-reinstall Lawrence R Kincheloe III 2025-11-23 02:19:33 -0600
• 68148b1bf3 Export TRITON_HIP_LLD_PATH in speedrun.sh for AMD ROCm google-labs-jules[bot] 2025-11-23 08:19:07 +0000
• a271eb0553
  Merge 861cbce2e9 into 4a87a0d19f Nitish Pandey 2025-11-23 12:22:12 +0530
• da035bf408
  Merge pull request #21 from LokiMetaSmith/fix-amd-triton-reinstall Lawrence R Kincheloe III 2025-11-23 00:49:55 -0600
• 1f9b734358 Use gloo backend for DDP on AMD ROCm to avoid NCCL crashes google-labs-jules[bot] 2025-11-23 06:49:07 +0000
• 861cbce2e9 fix condition to perform bpb evaluation Nitish Pandey 2025-11-23 12:07:40 +0530
• c1fc4400b0
  Merge pull request #20 from LokiMetaSmith/fix-amd-triton-reinstall Lawrence R Kincheloe III 2025-11-22 23:34:51 -0600
• 962deeefb6 Fix HIP invalid device ordinal error on multi-GPU setup google-labs-jules[bot] 2025-11-23 05:34:20 +0000
• 23695f817d
  Merge pull request #19 from LokiMetaSmith/fix-amd-triton-reinstall Lawrence R Kincheloe III 2025-11-22 23:22:57 -0600
• b92647c580 Fix AMD Triton runtime error by reinstalling pytorch-triton-rocm google-labs-jules[bot] 2025-11-23 05:22:21 +0000
• 44476a3512
  Merge pull request #18 from LokiMetaSmith/fix-amd-triton-reinstall Lawrence R Kincheloe III 2025-11-22 22:26:58 -0600
• d291a62ad8 Fix AMD Triton re-installation issue in speedrun.sh google-labs-jules[bot] 2025-11-23 04:26:32 +0000
• 054394c708
  Merge pull request #17 from LokiMetaSmith/amd-triton-fix Lawrence R Kincheloe III 2025-11-22 21:51:45 -0600
• 994491b28d Move triton uninstall after uv sync in speedrun.sh for AMD google-labs-jules[bot] 2025-11-23 03:50:46 +0000
• 33b6b800fa
  Merge pull request #16 from LokiMetaSmith/fix-amd-install Lawrence R Kincheloe III 2025-11-22 21:40:05 -0600
• 8881ea84bf Fix AMD Triton conflict in speedrun.sh google-labs-jules[bot] 2025-11-23 03:38:56 +0000
• d46e9a72d4
  Merge pull request #15 from LokiMetaSmith/fix-amd-install Lawrence R Kincheloe III 2025-11-22 20:33:32 -0600
• 83bb650b49 Fix AMD ROCm install regression in speedrun.sh google-labs-jules[bot] 2025-11-23 02:33:07 +0000
• dd37f29fe4
  Update Python version and torch dependencies Lawrence R Kincheloe III 2025-11-22 20:02:59 -0600
• 1af926205d
  Update Python version from 3.10 to 3.12 Lawrence R Kincheloe III 2025-11-22 20:00:57 -0600
• ddc51d34df
  Merge pull request #14 from LokiMetaSmith/fix-cpu-ddp-init Lawrence R Kincheloe III 2025-11-22 17:52:07 -0600
• 083de95913 Fix hardware detection for AMD ROCm and single-process CPU crashes google-labs-jules[bot] 2025-11-22 23:50:50 +0000
• b23494d2e2
  Update .gitignore Lawrence R Kincheloe III 2025-11-22 17:04:27 -0600
• 0fb04c8f25
  Merge 26b0941f75 into 4a87a0d19f Pyry Takala 2025-11-22 22:51:48 +0100
• 26b0941f75
  fix Sofie Van Landeghem 2025-11-22 22:51:42 +0100
• c09b897601
  further cleanup Sofie Van Landeghem 2025-11-22 22:50:34 +0100
• df9a644e24
  make code bit more succinct Sofie Van Landeghem 2025-11-22 22:48:55 +0100
• f5c5f8e055
  Merge 81a958350a into 4a87a0d19f Pyry Takala 2025-11-22 22:38:29 +0100
• 81a958350a
  do int conversation before taking the max Sofie Van Landeghem 2025-11-22 22:38:23 +0100
• 3b3113c8d2
  Merge pull request #13 from LokiMetaSmith/fix-cpu-ddp-init Lawrence R Kincheloe III 2025-11-22 12:07:29 -0600
• 28ef4c528e Fix CPU DDP crashes, enable ROCm detection, and prevent single-process distributed optimizer errors google-labs-jules[bot] 2025-11-22 18:06:58 +0000
• dfecef47bb fix last_step cal bug kuizhiqing 2025-11-23 00:04:55 +0800
• 36df08a5a9
  Merge pull request #12 from LokiMetaSmith/fix-cpu-ddp-init Lawrence R Kincheloe III 2025-11-22 03:19:05 -0600
• 48e632245e Fix ROCm/APU detection and CPU DDP OOM crash google-labs-jules[bot] 2025-11-22 09:18:40 +0000
• 8009354739
  Merge pull request #11 from LokiMetaSmith/fix-cpu-ddp-init Lawrence R Kincheloe III 2025-11-22 01:35:48 -0600
• a35621e726 Fix CPU DDP crashes: Init Gloo backend, prevent OOM by reducing NPROC, add script safety google-labs-jules[bot] 2025-11-22 05:31:47 +0000
• b984cda2e4
  Merge 53b3a4fb81 into 4a87a0d19f Hao Yuan 2025-11-22 11:24:13 +0800
• 53b3a4fb81 fix: missing val_bpb on resume Sanzo00 2025-11-22 11:04:20 +0800
• b5fd54ac1c
  Merge pull request #10 from LokiMetaSmith/fix-cpu-ddp-init Lawrence R Kincheloe III 2025-11-21 17:42:06 -0600
• 9235fe4000 Fix process group initialization for CPU DDP and improve cleanup safety google-labs-jules[bot] 2025-11-21 23:41:34 +0000
• 958543fcd5
  Update nanochat/checkpoint_manager.py Pyry Takala 2025-11-21 13:47:22 -0800
• a33d04dca1 Cap stop parameter and warn once when it exceeds dataset size Pyry Takala 2025-11-21 20:51:46 +0000
• 85e49943ed Gracefully handle stop > dataset_size with warning Pyry Takala 2025-11-21 20:04:33 +0000
• 3e2a0668b2 Refactor find_last_step to use os.listdir with regex filtering Pyry Takala 2025-11-21 19:21:01 +0000
• 2311d38e3f
  Merge 4bcc3bb698 into 4a87a0d19f ericsilberstein1 2025-11-21 13:19:50 +0100
• 4bcc3bb698 clarify comment svlandeg 2025-11-21 13:19:45 +0100
• 104308cf78
  Merge pull request #9 from LokiMetaSmith/fix-dataloader-typeerror Lawrence R Kincheloe III 2025-11-20 23:12:47 -0600