项目文件夹 工单

mirrors / karpathy--llm.c

项目文件夹 工单

提交

提交图

选择分支

隐藏合并请求

ademeure-layernormbf16-3

attempt_fix_zero1

feature/bias_backward2

feature/cublaslt

feature/cudnn

feature/delete_backward_act

feature/finetune_llama31py

feature/fix_int_overflow

feature/fp32_weight_master_copy

feature/gpt3v1

feature/grad_accum

feature/grad_accum3

feature/hellaswag

feature/hellaswagc

feature/hf_eval

feature/libdir

feature/load_bf16

feature/lr_schedulers

feature/managed2

feature/nccl-only

feature/openmpi_flexible_path

feature/organize_data

feature/outliers

feature/overlap-backward-reduce

feature/packed128

feature/permuted_dataloader

feature/pytorch_parity

feature/random_init

feature/recompute

feature/refactor_part2

feature/restore_from_master

feature/resume

feature/rolling_checkpoints

feature/sharded_data

feature/streams

feature/weight_decay_2d_only

feature/wikitext103

feature/write_checkpoints

fix/illegalaccess

fix/reshuffle_zero_grad

gordicaleksa-fix_dataloader2

llama3

main

master

move/attention

move/encoder_kernel

move/fused_classifier

move/gelu

move/matmul

overfit-batch-actually

refactor/code-to-zerocuh

separate_tokenizer

try/move_cudnn_to_llmc

黑白 彩色

• 8397fb1512 Free the memory in layernorm.c Marco Vinciguerra 2024-04-09 20:14:18 +02:00
• 3a4d783633 amend lancer 2024-04-09 09:59:03 -07:00
• 73c92912b0 fix error: identifier M_PI is undefined lancer 2024-04-09 09:51:28 -07:00
• 03f37cfb6b naive attention kernel only parallelizing over batch,time,heads. have to speed this up a lot Andrej Karpathy 2024-04-09 14:53:18 +00:00
• 41d5f56278 might as well push the few kernels that i feel ok about so far Andrej Karpathy 2024-04-09 14:34:13 +00:00
• 4d98c44904 Merge pull request #23 from DominguesAddem1974/fix/torch-warning Andrej 2024-04-09 05:45:18 -07:00
• daf1dd81b2 fix: torch warning of python demo DominguesAddem1974 2024-04-09 18:17:04 +08:00
• 69c1f179a8 Fix a typo Varun 2024-04-09 15:02:33 +05:30
• a22c22b1ee Merge pull request #14 from Ricardicus/master Andrej 2024-04-08 21:29:47 -07:00
• 5d9ed0cc70 Merge pull request #15 from lancerts/layer_norm_code_ptr Andrej 2024-04-08 21:28:04 -07:00
• f6b93912b7 minor clarifications karpathy 2024-04-08 21:24:34 -07:00
• 1df7bceff9 using the compiler at hand Rickard Hallerbäck 2024-04-09 05:37:11 +02:00
• 0d30f54d9a more notes on training/inference karpathy 2024-04-08 21:20:28 -07:00
• eecb234698 Include the pytorch layer_norm.cpp and layer_norm_kernel.cu code pointer for people who are interested in details lancer 2024-04-08 21:07:28 -07:00
• e4d251bad4 add comment on why we even have to implement layernorm forward pass in pytorch karpathy 2024-04-08 17:03:40 -07:00
• fbc879a83c fix hallucination of github copilot lol karpathy 2024-04-08 16:48:48 -07:00
• ce3fef00fb add clarification on checkpointing karpathy 2024-04-08 16:46:25 -07:00
• a99b86f2ef oops fix latex expression karpathy 2024-04-08 16:30:02 -07:00
• c39de59168 attach a simple tutorial karpathy 2024-04-08 16:26:38 -07:00
• 3bcb9ba7d2 remove unused variables karpathy 2024-04-08 15:10:51 -07:00
• b0fc807bd7 Merge pull request #2 from onuralpszr/py-tqdm Andrej 2024-04-08 14:42:56 -07:00
• 90ac963826 Merge branch 'master' into py-tqdm Andrej 2024-04-08 14:42:47 -07:00
• 40221ccff5 Merge pull request #4 from Ricardicus/master Andrej 2024-04-08 14:41:43 -07:00
• 0c2d8d771d Merge pull request #5 from alexziskind1/master Andrej 2024-04-08 14:39:35 -07:00
• b8f7b1c201 fix: add transformers to requirements.txt Alex Ziskind 2024-04-08 16:50:59 -04:00
• b2228d0e83 Minor correction for openmp check Rickard Hallerbäck 2024-04-08 22:33:47 +02:00
• 8465847284 fix: add tqdm to requirements.txt Onuralp SEZER 2024-04-08 23:26:14 +03:00
• 2790d263a6 oops fix readme bug karpathy 2024-04-08 12:42:44 -07:00
• e8e1628632 first commit of just the reference cpu fp32 gpt2 training karpathy 2024-04-08 12:41:21 -07:00