karpathy/llama2.c

Issues

• EOS

  #544 opened 12 days ago by wendadawen
  0

• Can this be compiled to run on Windows 10, or Windows XP?

  #516 opened 8 months ago by MartialTerran
  4

• runq.c quantization not symmetric

  #542 opened 2 months ago by hafezmg48
  0

• how to use train.py to train from llama2-HF model.

  #541 opened 2 months ago by bailuan
  0

• llama2.c calculation in FP16

  #539 opened 3 months ago by ztachip
  0

• Convert GGUF file format to llam2.c 's bin file format

  #538 opened 3 months ago by ztachip
  0

• Adding support to Llama 3.1

  #537 opened 4 months ago by asmit203
  1

• Everyone, I have implemented multi-token prediction of InfiniAttention and meta.

  #517 opened 4 months ago by win10ogod
  0

• [Suggestion] Enable Discussion

  #492 opened 4 months ago by trholding
  1

• Everyone, I have implemented multi-token prediction of InfiniAttention and meta.

  #518 opened 5 months ago by win10ogod
  3

• Runing llama2.c on a microcontroller

  #480 opened 5 months ago by mahdi259
  2

• Error with torch not compiled with cuda enabled

  #531 opened 6 months ago by berlinbrown
  1

• Fail for execute ./run with meta llama2_7b.bin

  #529 opened 6 months ago by sss28072637
  0

• failed to convert llama_2 7B model in .gguf to .bin format

  #527 opened 6 months ago by adi-lb-phoenix
  2

• Weight share of input and output embedding

  #526 opened 6 months ago by fangzhangmnm
  0

• Could llama2.c be adapted to BitNet?

  #500 opened 9 months ago by izaxon
  1

• Missing Sampler when running on multiple GPUs using DDP

  #522 opened 7 months ago by banyan-god
  3

• Training Tiny Stories: 'CUDA' -vs- 'MPS'

  #461 opened a year ago by dbl001
  2

• Simplified llama2.c.dll

  #507 opened 8 months ago by JohnClaw
  4

• Not an issue: Asking for help

  #520 opened 7 months ago by Hjertesvikt
  2

• support for SIMD in matmul might increase performance i think so...

  #458 opened 7 months ago by guna-sd
  1

• mmap failed! ./run llama2_7b_q80.bin

  #519 opened 7 months ago by codetown
  0

• the export model and read_checkpoint is conflict

  #501 opened 9 months ago by l1351868270
  2

• Tokenizer errors out when inferencing llama2

  #502 opened 8 months ago by navidsam
  1

• malloc failed! on stories260 model

  #510 opened 8 months ago by vikramdattu
  1

• How about Llama3？

  #506 opened 8 months ago by franktaTian
  1

• Can the Huggingface model be converted to ckpt.pt to support training?

  #505 opened 8 months ago by win10ogod
  0

• RuntimeError with CUDA assertion failure when resuming model training from checkpoint

  #499 opened 9 months ago by fancling
  1

• add feature: export (quantize) from Llama2.c format

  #496 opened 9 months ago by hafezmg48
  0

• Once upon a time, there was a little girl named Lily

  #465 opened a year ago by GilesBathgate
  5

• Can you make a sora (diffusion transformer) tutorial similar to llama2.c?

  #489 opened 10 months ago by win10ogod
  1

• I'm doing an experiment with image generation, but my script outputs a binary file, how can I train a model using llama2.c?

  #488 opened 10 months ago by win10ogod
  0

• Could anyone port deepseek-moe to llama2.c?

  #485 opened 10 months ago by win10ogod
  0

• Please implement a project

  #484 opened 10 months ago by win10ogod
  0

• New Visual Walkthrough of Llama2.c

  #483 opened 10 months ago by ZoroDerVonCodier
  0

• Mobile React native Support Ported

  #482 opened 10 months ago by hootan09
  0

• Understanding "multiple_of"

  #481 opened 10 months ago by akbayt
  0

• Train/val split

  #478 opened a year ago by DavidHerel
  0

• Code/script to reproduce val loss using the shared models

  #475 opened a year ago by Alexey234432
  3

• How to quantize stories15M.bin

  #476 opened a year ago by forcekeng
  1

• can i train on cpu

  #456 opened a year ago by pure-water
  5

• HOw to add different coropus ?

  #462 opened a year ago by pure-water
  4

• Keras based tiny llama implementations

  #472 opened a year ago by Alexey234432
  0

• casual attention implementation

  #469 opened a year ago by liecn
  3

• How do we reproduce your stories, by a more practical Q&A chat model?

  #467 opened a year ago by oliverbob
  1

• Llama-shepherd-cli a small tool to keep track of implementations in various languages

  #471 opened a year ago by mikepapadim
  0

• NanoGPT in c for inference

  #464 opened a year ago by liecn
  0

• export does not seem to work?

  #454 opened a year ago by pure-water
  2

• mfu calculation

  #460 opened a year ago by pure-water
  0

• Is it possible to use Orca2 with this code ?

  #452 opened a year ago by FiveTechSoft
  1