Upload 6 files
Browse files- config_pretrain.json +40 -0
- config_sft.json +40 -0
- nano_168m_307000.pt +3 -0
- nano_168m_307000_sft_379000.pt +3 -0
- nano_168m_320000.pt +3 -0
- sft.log +0 -0
config_pretrain.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"use_lora": false,
|
| 3 |
+
"lora_rank": 8,
|
| 4 |
+
"lora_alpha": 16,
|
| 5 |
+
"lora_dropout": 0.0,
|
| 6 |
+
|
| 7 |
+
"from_checkpoint": "/home/bd4sur/ai/Nano/checkpoint/checkpoint_20241122_232939_step_196000.pt",
|
| 8 |
+
"save_checkpoint_to": "/home/bd4sur/ai/Nano/checkpoint",
|
| 9 |
+
"dataset_path": [
|
| 10 |
+
["/home/bd4sur/ai/Nano/dataset_preprocessed/pt_train_0.base64", "/home/bd4sur/ai/Nano/dataset_preprocessed/pt_val_0.base64"]
|
| 11 |
+
],
|
| 12 |
+
"tokenizer_path": "/home/bd4sur/ai/Nano/tokenizer/tokenizer_16384.json",
|
| 13 |
+
|
| 14 |
+
"random_seed": 39,
|
| 15 |
+
"batch_size": 80,
|
| 16 |
+
"gradient_accumulation_steps": 1,
|
| 17 |
+
"grad_clip": 1.0,
|
| 18 |
+
|
| 19 |
+
"dropout": 0.0,
|
| 20 |
+
|
| 21 |
+
"learning_rate": 5e-4,
|
| 22 |
+
"weight_decay": 1e-1,
|
| 23 |
+
"beta1": 0.9,
|
| 24 |
+
"beta2": 0.95,
|
| 25 |
+
|
| 26 |
+
"decay_lr": true,
|
| 27 |
+
"warmup_iters": 500,
|
| 28 |
+
"lr_decay_iters": 1e9,
|
| 29 |
+
"min_lr": 6e-5,
|
| 30 |
+
|
| 31 |
+
"eval_interval": 500,
|
| 32 |
+
"log_interval": 10,
|
| 33 |
+
"eval_iters": 2,
|
| 34 |
+
|
| 35 |
+
"backend": "nccl",
|
| 36 |
+
"device": "cuda",
|
| 37 |
+
"sdp_kernel": "flash",
|
| 38 |
+
"dtype": "bfloat16",
|
| 39 |
+
"use_amp": true
|
| 40 |
+
}
|
config_sft.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"use_lora": false,
|
| 3 |
+
"lora_rank": 8,
|
| 4 |
+
"lora_alpha": 16,
|
| 5 |
+
"lora_dropout": 0.0,
|
| 6 |
+
|
| 7 |
+
"from_checkpoint": "/root/autodl-tmp/checkpoint_20241130_005942_step_307000.pt",
|
| 8 |
+
"save_checkpoint_to": "/root/autodl-tmp/checkpoint",
|
| 9 |
+
"dataset_path": [
|
| 10 |
+
["/root/autodl-tmp/sft_train.base64", "/root/autodl-tmp/sft_val.base64"]
|
| 11 |
+
],
|
| 12 |
+
"tokenizer_path": "/root/Nano/tokenizer/tokenizer_16384.json",
|
| 13 |
+
|
| 14 |
+
"random_seed": 39,
|
| 15 |
+
"batch_size": 120,
|
| 16 |
+
"gradient_accumulation_steps": 2,
|
| 17 |
+
"grad_clip": 1.0,
|
| 18 |
+
|
| 19 |
+
"dropout": 0.1,
|
| 20 |
+
|
| 21 |
+
"learning_rate": 1e-6,
|
| 22 |
+
"weight_decay": 1e-1,
|
| 23 |
+
"beta1": 0.9,
|
| 24 |
+
"beta2": 0.95,
|
| 25 |
+
|
| 26 |
+
"decay_lr": false,
|
| 27 |
+
"warmup_iters": 10000,
|
| 28 |
+
"lr_decay_iters": 1e9,
|
| 29 |
+
"min_lr": 1e-8,
|
| 30 |
+
|
| 31 |
+
"eval_interval": 500,
|
| 32 |
+
"log_interval": 10,
|
| 33 |
+
"eval_iters": 5,
|
| 34 |
+
|
| 35 |
+
"backend": "nccl",
|
| 36 |
+
"device": "cuda",
|
| 37 |
+
"sdp_kernel": "flash",
|
| 38 |
+
"dtype": "bfloat16",
|
| 39 |
+
"use_amp": true
|
| 40 |
+
}
|
nano_168m_307000.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac4e4469d8346cb01195bf804115a2693b548610729ecb1742de51ef81ffaa29
|
| 3 |
+
size 2045994798
|
nano_168m_307000_sft_379000.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c184f0373682a4f2715163464358b0a8882e392a7fed6dd9f24563aafe45747
|
| 3 |
+
size 2045994734
|
nano_168m_320000.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:162bc16899a55900f044c656146e3687b3260e9c295a9e4b79e5cebd6e4d960d
|
| 3 |
+
size 2045994798
|
sft.log
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|