Finish pretraining 15_000 steps phase 1

Finish pretraining 13k steps phase 1

Finish pretraining 12k steps phase 1

Finish pretraining 10_000 steps phase 1

Finish pretraining 4k steps phase 1

Finish pretraining 2k steps phase 1

Finish finetuning 5k steps phase 1 and 5k steps phase 2 start from pretrained weights

Finish finetuning 5k steps phase 1 and 1k steps phase 2 start from pretrained weights

Finish finetuning 5000 steps for phase 1 start from pretrained weights

Finish 2000 steps for finetuned model start from pretrained weights

Finish training for 15k steps for phase 1 and 15k steps for phase 2

delete old checkpoints

delete old checkpoints

delete old checkpoints

delete old checkpoints

delete old checkpoints

delete old checkpoints

Finish training for 15k steps for phase 1 and 12.5k steps for phase 2

Delete phase1/checkpoint_step_3000

Delete phase1/final

Finish training for 12k steps for phase 1 and 10k steps for phase 2

Finish training for 12k steps for phase 1 and 10k steps for phase 2

Finish training for 10k steps for phase 1 and 10k steps for phase 2

Finish training for 10k steps for phase 1 and 9k steps for phase 2

Finish training for 10k steps for phase 1 and 7k steps for phase 2

Finish training for 10k steps for phase 1 and 5k steps for phase 2

Delete phase2/checkpoint_step_2

Delete phase2/checkpoint_step_1

Add converted safetensors and original pth weights

Update README.md

Update README.md

Add converted safetensors and original pth weights

Add converted safetensors and original pth weights

Delete files global* with huggingface_hub

Add converted safetensors and original pth weights

Add converted safetensors and original pth weights

Delete global_step_250

Add converted safetensors and original pth weights

Add converted safetensors and original pth weights

Add converted safetensors and original pth weights

Add converted safetensors and original pth weights

Add converted safetensors and original pth weights

Add converted safetensors and original pth weights

ensure this the pretrained model of the paper

Add converted safetensors and original pth weights

Update README.md

Update README.md

Create README.md

Add converted safetensors and original pth weights

initial commit