Upload 5 files
Browse files- additional_config.json +1 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1111 -0
- training_args.bin +3 -0
additional_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"lora_dtype": null, "lorap_lr_ratio": null, "lorap_emb_lr": 1e-06}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69f50a692634404f2eebb2eab9f456865957578d752987bc52d843ac2a774366
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe30a5209d869e3db5dba9d278dc289ca2cfdb48b52e96cb3f9daf3ed899c3bd
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1111 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.51153946,
|
| 3 |
+
"best_model_checkpoint": "/export/home2/zli/kc/mm_rag/Qwen2.5-72B-Instruct_lora/checkpoint-513",
|
| 4 |
+
"epoch": 0.9995129079396006,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 513,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.001948368241597662,
|
| 13 |
+
"grad_norm": 0.7911198735237122,
|
| 14 |
+
"learning_rate": 3.846153846153847e-06,
|
| 15 |
+
"loss": 1.1104155778884888,
|
| 16 |
+
"memory(GiB)": 239.17,
|
| 17 |
+
"step": 1,
|
| 18 |
+
"token_acc": 0.7912871287128713,
|
| 19 |
+
"train_speed(iter/s)": 0.00699
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.00974184120798831,
|
| 23 |
+
"grad_norm": 1.048244833946228,
|
| 24 |
+
"learning_rate": 1.923076923076923e-05,
|
| 25 |
+
"loss": 1.0879460573196411,
|
| 26 |
+
"memory(GiB)": 257.56,
|
| 27 |
+
"step": 5,
|
| 28 |
+
"token_acc": 0.7876893738743511,
|
| 29 |
+
"train_speed(iter/s)": 0.006513
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.01948368241597662,
|
| 33 |
+
"grad_norm": 0.8897583484649658,
|
| 34 |
+
"learning_rate": 3.846153846153846e-05,
|
| 35 |
+
"loss": 1.179165744781494,
|
| 36 |
+
"memory(GiB)": 267.89,
|
| 37 |
+
"step": 10,
|
| 38 |
+
"token_acc": 0.7578167359479486,
|
| 39 |
+
"train_speed(iter/s)": 0.006891
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.029225523623964928,
|
| 43 |
+
"grad_norm": 0.3992486298084259,
|
| 44 |
+
"learning_rate": 5.769230769230769e-05,
|
| 45 |
+
"loss": 0.866185188293457,
|
| 46 |
+
"memory(GiB)": 267.89,
|
| 47 |
+
"step": 15,
|
| 48 |
+
"token_acc": 0.7692150449713818,
|
| 49 |
+
"train_speed(iter/s)": 0.006796
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.03896736483195324,
|
| 53 |
+
"grad_norm": 0.50007164478302,
|
| 54 |
+
"learning_rate": 7.692307692307693e-05,
|
| 55 |
+
"loss": 0.7950119972229004,
|
| 56 |
+
"memory(GiB)": 267.89,
|
| 57 |
+
"step": 20,
|
| 58 |
+
"token_acc": 0.7725610358802016,
|
| 59 |
+
"train_speed(iter/s)": 0.006865
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.04870920603994155,
|
| 63 |
+
"grad_norm": 0.2926434874534607,
|
| 64 |
+
"learning_rate": 9.615384615384617e-05,
|
| 65 |
+
"loss": 0.6839358806610107,
|
| 66 |
+
"memory(GiB)": 267.89,
|
| 67 |
+
"step": 25,
|
| 68 |
+
"token_acc": 0.7947733918128655,
|
| 69 |
+
"train_speed(iter/s)": 0.006974
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.058451047247929856,
|
| 73 |
+
"grad_norm": 0.4686521887779236,
|
| 74 |
+
"learning_rate": 9.998335523311734e-05,
|
| 75 |
+
"loss": 0.6574328422546387,
|
| 76 |
+
"memory(GiB)": 267.89,
|
| 77 |
+
"step": 30,
|
| 78 |
+
"token_acc": 0.8028529939352524,
|
| 79 |
+
"train_speed(iter/s)": 0.006948
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.06819288845591817,
|
| 83 |
+
"grad_norm": 0.2973114252090454,
|
| 84 |
+
"learning_rate": 9.991575486016592e-05,
|
| 85 |
+
"loss": 0.6306457042694091,
|
| 86 |
+
"memory(GiB)": 267.89,
|
| 87 |
+
"step": 35,
|
| 88 |
+
"token_acc": 0.8055379603655778,
|
| 89 |
+
"train_speed(iter/s)": 0.006928
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.07793472966390648,
|
| 93 |
+
"grad_norm": 0.2803875803947449,
|
| 94 |
+
"learning_rate": 9.979622885143301e-05,
|
| 95 |
+
"loss": 0.6483866691589355,
|
| 96 |
+
"memory(GiB)": 267.89,
|
| 97 |
+
"step": 40,
|
| 98 |
+
"token_acc": 0.8065050756163248,
|
| 99 |
+
"train_speed(iter/s)": 0.006967
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.08767657087189479,
|
| 103 |
+
"grad_norm": 0.486200213432312,
|
| 104 |
+
"learning_rate": 9.962490154569727e-05,
|
| 105 |
+
"loss": 0.6592767715454102,
|
| 106 |
+
"memory(GiB)": 279.45,
|
| 107 |
+
"step": 45,
|
| 108 |
+
"token_acc": 0.7944840703756538,
|
| 109 |
+
"train_speed(iter/s)": 0.006889
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.0974184120798831,
|
| 113 |
+
"grad_norm": 0.3190113604068756,
|
| 114 |
+
"learning_rate": 9.940195116883755e-05,
|
| 115 |
+
"loss": 0.6488125324249268,
|
| 116 |
+
"memory(GiB)": 279.45,
|
| 117 |
+
"step": 50,
|
| 118 |
+
"token_acc": 0.7989409984871407,
|
| 119 |
+
"train_speed(iter/s)": 0.006919
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.1071602532878714,
|
| 123 |
+
"grad_norm": 0.2820813059806824,
|
| 124 |
+
"learning_rate": 9.91276096484306e-05,
|
| 125 |
+
"loss": 0.6796814918518066,
|
| 126 |
+
"memory(GiB)": 279.45,
|
| 127 |
+
"step": 55,
|
| 128 |
+
"token_acc": 0.7929543782192788,
|
| 129 |
+
"train_speed(iter/s)": 0.006811
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.11690209449585971,
|
| 133 |
+
"grad_norm": 0.37464168667793274,
|
| 134 |
+
"learning_rate": 9.880216237248481e-05,
|
| 135 |
+
"loss": 0.7066696166992188,
|
| 136 |
+
"memory(GiB)": 279.45,
|
| 137 |
+
"step": 60,
|
| 138 |
+
"token_acc": 0.7865912142720123,
|
| 139 |
+
"train_speed(iter/s)": 0.006893
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.12664393570384802,
|
| 143 |
+
"grad_norm": 0.3829931318759918,
|
| 144 |
+
"learning_rate": 9.842594789256103e-05,
|
| 145 |
+
"loss": 0.5756209850311279,
|
| 146 |
+
"memory(GiB)": 279.45,
|
| 147 |
+
"step": 65,
|
| 148 |
+
"token_acc": 0.8192581553237291,
|
| 149 |
+
"train_speed(iter/s)": 0.0069
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.13638577691183634,
|
| 153 |
+
"grad_norm": 0.3420902490615845,
|
| 154 |
+
"learning_rate": 9.799935757158891e-05,
|
| 155 |
+
"loss": 0.7137881755828858,
|
| 156 |
+
"memory(GiB)": 279.45,
|
| 157 |
+
"step": 70,
|
| 158 |
+
"token_acc": 0.7774079467443312,
|
| 159 |
+
"train_speed(iter/s)": 0.006961
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.14612761811982464,
|
| 163 |
+
"grad_norm": 0.2829829454421997,
|
| 164 |
+
"learning_rate": 9.752283517674575e-05,
|
| 165 |
+
"loss": 0.5826292037963867,
|
| 166 |
+
"memory(GiB)": 279.45,
|
| 167 |
+
"step": 75,
|
| 168 |
+
"token_acc": 0.8175895765472313,
|
| 169 |
+
"train_speed(iter/s)": 0.006986
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.15586945932781296,
|
| 173 |
+
"grad_norm": 0.4237753748893738,
|
| 174 |
+
"learning_rate": 9.699687641782067e-05,
|
| 175 |
+
"loss": 0.5978005886077881,
|
| 176 |
+
"memory(GiB)": 279.45,
|
| 177 |
+
"step": 80,
|
| 178 |
+
"token_acc": 0.8125563774129533,
|
| 179 |
+
"train_speed(iter/s)": 0.007002
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.16561130053580125,
|
| 183 |
+
"grad_norm": 0.6381401419639587,
|
| 184 |
+
"learning_rate": 9.642202843154491e-05,
|
| 185 |
+
"loss": 0.6211316585540771,
|
| 186 |
+
"memory(GiB)": 279.45,
|
| 187 |
+
"step": 85,
|
| 188 |
+
"token_acc": 0.8153757510906248,
|
| 189 |
+
"train_speed(iter/s)": 0.007036
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.17535314174378958,
|
| 193 |
+
"grad_norm": 0.3631569743156433,
|
| 194 |
+
"learning_rate": 9.579888921242439e-05,
|
| 195 |
+
"loss": 0.6291302680969239,
|
| 196 |
+
"memory(GiB)": 279.45,
|
| 197 |
+
"step": 90,
|
| 198 |
+
"token_acc": 0.8056357749190514,
|
| 199 |
+
"train_speed(iter/s)": 0.00706
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.1850949829517779,
|
| 203 |
+
"grad_norm": 0.31778669357299805,
|
| 204 |
+
"learning_rate": 9.512810699066667e-05,
|
| 205 |
+
"loss": 0.6138756752014161,
|
| 206 |
+
"memory(GiB)": 279.45,
|
| 207 |
+
"step": 95,
|
| 208 |
+
"token_acc": 0.8115504301665751,
|
| 209 |
+
"train_speed(iter/s)": 0.007048
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.1948368241597662,
|
| 213 |
+
"grad_norm": 0.3409082889556885,
|
| 214 |
+
"learning_rate": 9.441037955784944e-05,
|
| 215 |
+
"loss": 0.640412425994873,
|
| 216 |
+
"memory(GiB)": 279.45,
|
| 217 |
+
"step": 100,
|
| 218 |
+
"token_acc": 0.7977538518656023,
|
| 219 |
+
"train_speed(iter/s)": 0.007066
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1948368241597662,
|
| 223 |
+
"eval_loss": 0.6026075482368469,
|
| 224 |
+
"eval_runtime": 26.8674,
|
| 225 |
+
"eval_samples_per_second": 0.149,
|
| 226 |
+
"eval_steps_per_second": 0.074,
|
| 227 |
+
"step": 100
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.20457866536775451,
|
| 231 |
+
"grad_norm": 0.5213446021080017,
|
| 232 |
+
"learning_rate": 9.364645354103206e-05,
|
| 233 |
+
"loss": 0.6617852210998535,
|
| 234 |
+
"memory(GiB)": 279.45,
|
| 235 |
+
"step": 105,
|
| 236 |
+
"token_acc": 0.7954326459818848,
|
| 237 |
+
"train_speed(iter/s)": 0.007023
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.2143205065757428,
|
| 241 |
+
"grad_norm": 0.3881484270095825,
|
| 242 |
+
"learning_rate": 9.28371236260652e-05,
|
| 243 |
+
"loss": 0.6231249809265137,
|
| 244 |
+
"memory(GiB)": 279.45,
|
| 245 |
+
"step": 110,
|
| 246 |
+
"token_acc": 0.8025908424311553,
|
| 247 |
+
"train_speed(iter/s)": 0.007042
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.22406234778373113,
|
| 251 |
+
"grad_norm": 0.34954214096069336,
|
| 252 |
+
"learning_rate": 9.198323173090663e-05,
|
| 253 |
+
"loss": 0.6497172355651856,
|
| 254 |
+
"memory(GiB)": 279.45,
|
| 255 |
+
"step": 115,
|
| 256 |
+
"token_acc": 0.794488631798827,
|
| 257 |
+
"train_speed(iter/s)": 0.007052
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.23380418899171942,
|
| 261 |
+
"grad_norm": 0.3073936402797699,
|
| 262 |
+
"learning_rate": 9.108566612980298e-05,
|
| 263 |
+
"loss": 0.6538199424743653,
|
| 264 |
+
"memory(GiB)": 279.45,
|
| 265 |
+
"step": 120,
|
| 266 |
+
"token_acc": 0.7920408356177332,
|
| 267 |
+
"train_speed(iter/s)": 0.00707
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.24354603019970775,
|
| 271 |
+
"grad_norm": 0.3140193819999695,
|
| 272 |
+
"learning_rate": 9.014536052924883e-05,
|
| 273 |
+
"loss": 0.5965206623077393,
|
| 274 |
+
"memory(GiB)": 279.45,
|
| 275 |
+
"step": 125,
|
| 276 |
+
"token_acc": 0.8178110129163834,
|
| 277 |
+
"train_speed(iter/s)": 0.007102
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 0.25328787140769604,
|
| 281 |
+
"grad_norm": 0.9085149168968201,
|
| 282 |
+
"learning_rate": 8.916329309668397e-05,
|
| 283 |
+
"loss": 0.5718743324279785,
|
| 284 |
+
"memory(GiB)": 279.45,
|
| 285 |
+
"step": 130,
|
| 286 |
+
"token_acc": 0.8312764304824373,
|
| 287 |
+
"train_speed(iter/s)": 0.007101
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.26302971261568436,
|
| 291 |
+
"grad_norm": 0.4713858664035797,
|
| 292 |
+
"learning_rate": 8.814048544293965e-05,
|
| 293 |
+
"loss": 0.582437801361084,
|
| 294 |
+
"memory(GiB)": 279.45,
|
| 295 |
+
"step": 135,
|
| 296 |
+
"token_acc": 0.8134561744569352,
|
| 297 |
+
"train_speed(iter/s)": 0.007088
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.2727715538236727,
|
| 301 |
+
"grad_norm": 0.39269959926605225,
|
| 302 |
+
"learning_rate": 8.707800155949217e-05,
|
| 303 |
+
"loss": 0.5708662986755371,
|
| 304 |
+
"memory(GiB)": 279.45,
|
| 305 |
+
"step": 140,
|
| 306 |
+
"token_acc": 0.8203411790110128,
|
| 307 |
+
"train_speed(iter/s)": 0.007075
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 0.282513395031661,
|
| 311 |
+
"grad_norm": 0.38881248235702515,
|
| 312 |
+
"learning_rate": 8.597694671162921e-05,
|
| 313 |
+
"loss": 0.6288607597351075,
|
| 314 |
+
"memory(GiB)": 279.45,
|
| 315 |
+
"step": 145,
|
| 316 |
+
"token_acc": 0.8012965630788228,
|
| 317 |
+
"train_speed(iter/s)": 0.007075
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.2922552362396493,
|
| 321 |
+
"grad_norm": 1.00093412399292,
|
| 322 |
+
"learning_rate": 8.483846628868055e-05,
|
| 323 |
+
"loss": 0.6310084819793701,
|
| 324 |
+
"memory(GiB)": 279.45,
|
| 325 |
+
"step": 150,
|
| 326 |
+
"token_acc": 0.8051169868746433,
|
| 327 |
+
"train_speed(iter/s)": 0.007102
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.3019970774476376,
|
| 331 |
+
"grad_norm": 0.5331721901893616,
|
| 332 |
+
"learning_rate": 8.366374461250916e-05,
|
| 333 |
+
"loss": 0.5937428951263428,
|
| 334 |
+
"memory(GiB)": 279.45,
|
| 335 |
+
"step": 155,
|
| 336 |
+
"token_acc": 0.8136619109381755,
|
| 337 |
+
"train_speed(iter/s)": 0.007062
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.3117389186556259,
|
| 341 |
+
"grad_norm": 0.6358391642570496,
|
| 342 |
+
"learning_rate": 8.245400370550198e-05,
|
| 343 |
+
"loss": 0.6008272171020508,
|
| 344 |
+
"memory(GiB)": 279.45,
|
| 345 |
+
"step": 160,
|
| 346 |
+
"token_acc": 0.8157670857358919,
|
| 347 |
+
"train_speed(iter/s)": 0.007051
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.32148075986361424,
|
| 351 |
+
"grad_norm": 0.40399816632270813,
|
| 352 |
+
"learning_rate": 8.121050201934235e-05,
|
| 353 |
+
"loss": 0.6029816627502441,
|
| 354 |
+
"memory(GiB)": 279.45,
|
| 355 |
+
"step": 165,
|
| 356 |
+
"token_acc": 0.8082507259437268,
|
| 357 |
+
"train_speed(iter/s)": 0.007056
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.3312226010716025,
|
| 361 |
+
"grad_norm": 0.4126909673213959,
|
| 362 |
+
"learning_rate": 7.993453312588607e-05,
|
| 363 |
+
"loss": 0.6517924785614013,
|
| 364 |
+
"memory(GiB)": 279.45,
|
| 365 |
+
"step": 170,
|
| 366 |
+
"token_acc": 0.7958906764876914,
|
| 367 |
+
"train_speed(iter/s)": 0.007056
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.34096444227959083,
|
| 371 |
+
"grad_norm": 0.6618629097938538,
|
| 372 |
+
"learning_rate": 7.862742437150336e-05,
|
| 373 |
+
"loss": 0.6141845226287842,
|
| 374 |
+
"memory(GiB)": 279.45,
|
| 375 |
+
"step": 175,
|
| 376 |
+
"token_acc": 0.8037235096058626,
|
| 377 |
+
"train_speed(iter/s)": 0.007045
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.35070628348757915,
|
| 381 |
+
"grad_norm": 0.5881444215774536,
|
| 382 |
+
"learning_rate": 7.729053549628622e-05,
|
| 383 |
+
"loss": 0.6001346588134766,
|
| 384 |
+
"memory(GiB)": 279.45,
|
| 385 |
+
"step": 180,
|
| 386 |
+
"token_acc": 0.8113048368953881,
|
| 387 |
+
"train_speed(iter/s)": 0.007061
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.3604481246955675,
|
| 391 |
+
"grad_norm": 1.692840576171875,
|
| 392 |
+
"learning_rate": 7.592525721955786e-05,
|
| 393 |
+
"loss": 0.574149227142334,
|
| 394 |
+
"memory(GiB)": 279.45,
|
| 395 |
+
"step": 185,
|
| 396 |
+
"token_acc": 0.8219587430113746,
|
| 397 |
+
"train_speed(iter/s)": 0.007067
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 0.3701899659035558,
|
| 401 |
+
"grad_norm": 0.38788875937461853,
|
| 402 |
+
"learning_rate": 7.45330097931553e-05,
|
| 403 |
+
"loss": 0.6197400093078613,
|
| 404 |
+
"memory(GiB)": 279.45,
|
| 405 |
+
"step": 190,
|
| 406 |
+
"token_acc": 0.8041193060422186,
|
| 407 |
+
"train_speed(iter/s)": 0.007079
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.37993180711154406,
|
| 411 |
+
"grad_norm": 0.48055437207221985,
|
| 412 |
+
"learning_rate": 7.311524152399054e-05,
|
| 413 |
+
"loss": 0.6016897678375244,
|
| 414 |
+
"memory(GiB)": 279.45,
|
| 415 |
+
"step": 195,
|
| 416 |
+
"token_acc": 0.8031721765717562,
|
| 417 |
+
"train_speed(iter/s)": 0.007068
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.3896736483195324,
|
| 421 |
+
"grad_norm": 0.4017133116722107,
|
| 422 |
+
"learning_rate": 7.167342726742685e-05,
|
| 423 |
+
"loss": 0.5860675811767578,
|
| 424 |
+
"memory(GiB)": 279.45,
|
| 425 |
+
"step": 200,
|
| 426 |
+
"token_acc": 0.8191920830629461,
|
| 427 |
+
"train_speed(iter/s)": 0.007059
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.3896736483195324,
|
| 431 |
+
"eval_loss": 0.5398527979850769,
|
| 432 |
+
"eval_runtime": 26.925,
|
| 433 |
+
"eval_samples_per_second": 0.149,
|
| 434 |
+
"eval_steps_per_second": 0.074,
|
| 435 |
+
"step": 200
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 0.3994154895275207,
|
| 439 |
+
"grad_norm": 0.48819786310195923,
|
| 440 |
+
"learning_rate": 7.020906689303766e-05,
|
| 441 |
+
"loss": 0.60306715965271,
|
| 442 |
+
"memory(GiB)": 279.45,
|
| 443 |
+
"step": 205,
|
| 444 |
+
"token_acc": 0.8102000173175167,
|
| 445 |
+
"train_speed(iter/s)": 0.007055
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.40915733073550903,
|
| 449 |
+
"grad_norm": 0.5116552114486694,
|
| 450 |
+
"learning_rate": 6.872368372434416e-05,
|
| 451 |
+
"loss": 0.6070918083190918,
|
| 452 |
+
"memory(GiB)": 279.45,
|
| 453 |
+
"step": 210,
|
| 454 |
+
"token_acc": 0.804848606780002,
|
| 455 |
+
"train_speed(iter/s)": 0.007061
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.4188991719434973,
|
| 459 |
+
"grad_norm": 0.42464399337768555,
|
| 460 |
+
"learning_rate": 6.721882295415425e-05,
|
| 461 |
+
"loss": 0.633366060256958,
|
| 462 |
+
"memory(GiB)": 279.45,
|
| 463 |
+
"step": 215,
|
| 464 |
+
"token_acc": 0.8031126254719587,
|
| 465 |
+
"train_speed(iter/s)": 0.007047
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.4286410131514856,
|
| 469 |
+
"grad_norm": 0.43728524446487427,
|
| 470 |
+
"learning_rate": 6.569605003715201e-05,
|
| 471 |
+
"loss": 0.5457853317260742,
|
| 472 |
+
"memory(GiB)": 279.45,
|
| 473 |
+
"step": 220,
|
| 474 |
+
"token_acc": 0.8268458881732474,
|
| 475 |
+
"train_speed(iter/s)": 0.007041
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 0.43838285435947394,
|
| 479 |
+
"grad_norm": 0.526282548904419,
|
| 480 |
+
"learning_rate": 6.41569490614092e-05,
|
| 481 |
+
"loss": 0.5758168220520019,
|
| 482 |
+
"memory(GiB)": 291.53,
|
| 483 |
+
"step": 225,
|
| 484 |
+
"token_acc": 0.8169302630500697,
|
| 485 |
+
"train_speed(iter/s)": 0.007022
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.44812469556746226,
|
| 489 |
+
"grad_norm": 0.4335802495479584,
|
| 490 |
+
"learning_rate": 6.260312110051312e-05,
|
| 491 |
+
"loss": 0.5597887992858886,
|
| 492 |
+
"memory(GiB)": 291.53,
|
| 493 |
+
"step": 230,
|
| 494 |
+
"token_acc": 0.8260340632603407,
|
| 495 |
+
"train_speed(iter/s)": 0.007017
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 0.4578665367754506,
|
| 499 |
+
"grad_norm": 0.46290066838264465,
|
| 500 |
+
"learning_rate": 6.103618254802511e-05,
|
| 501 |
+
"loss": 0.5268624782562256,
|
| 502 |
+
"memory(GiB)": 291.53,
|
| 503 |
+
"step": 235,
|
| 504 |
+
"token_acc": 0.8337082864641919,
|
| 505 |
+
"train_speed(iter/s)": 0.007024
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.46760837798343885,
|
| 509 |
+
"grad_norm": 0.44755905866622925,
|
| 510 |
+
"learning_rate": 5.945776343600207e-05,
|
| 511 |
+
"loss": 0.6087980270385742,
|
| 512 |
+
"memory(GiB)": 291.53,
|
| 513 |
+
"step": 240,
|
| 514 |
+
"token_acc": 0.8013279748383715,
|
| 515 |
+
"train_speed(iter/s)": 0.007003
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.47735021919142717,
|
| 519 |
+
"grad_norm": 0.35961049795150757,
|
| 520 |
+
"learning_rate": 5.7869505739330546e-05,
|
| 521 |
+
"loss": 0.6320755958557129,
|
| 522 |
+
"memory(GiB)": 291.53,
|
| 523 |
+
"step": 245,
|
| 524 |
+
"token_acc": 0.7990379032926931,
|
| 525 |
+
"train_speed(iter/s)": 0.006988
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.4870920603994155,
|
| 529 |
+
"grad_norm": 0.43942877650260925,
|
| 530 |
+
"learning_rate": 5.627306166763684e-05,
|
| 531 |
+
"loss": 0.5911942958831787,
|
| 532 |
+
"memory(GiB)": 291.53,
|
| 533 |
+
"step": 250,
|
| 534 |
+
"token_acc": 0.8108974358974359,
|
| 535 |
+
"train_speed(iter/s)": 0.006984
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.4968339016074038,
|
| 539 |
+
"grad_norm": 0.3216777443885803,
|
| 540 |
+
"learning_rate": 5.467009194655045e-05,
|
| 541 |
+
"loss": 0.5380321025848389,
|
| 542 |
+
"memory(GiB)": 291.53,
|
| 543 |
+
"step": 255,
|
| 544 |
+
"token_acc": 0.8256673332753673,
|
| 545 |
+
"train_speed(iter/s)": 0.006954
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.5065757428153921,
|
| 549 |
+
"grad_norm": 0.3348521590232849,
|
| 550 |
+
"learning_rate": 5.306226409010855e-05,
|
| 551 |
+
"loss": 0.5896577358245849,
|
| 552 |
+
"memory(GiB)": 291.53,
|
| 553 |
+
"step": 260,
|
| 554 |
+
"token_acc": 0.8136957057229658,
|
| 555 |
+
"train_speed(iter/s)": 0.00696
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.5163175840233805,
|
| 559 |
+
"grad_norm": 0.4316939115524292,
|
| 560 |
+
"learning_rate": 5.145125066609877e-05,
|
| 561 |
+
"loss": 0.5456296920776367,
|
| 562 |
+
"memory(GiB)": 291.53,
|
| 563 |
+
"step": 265,
|
| 564 |
+
"token_acc": 0.8252879627187197,
|
| 565 |
+
"train_speed(iter/s)": 0.006968
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 0.5260594252313687,
|
| 569 |
+
"grad_norm": 0.4729405343532562,
|
| 570 |
+
"learning_rate": 4.983872755614461e-05,
|
| 571 |
+
"loss": 0.5579260349273681,
|
| 572 |
+
"memory(GiB)": 291.53,
|
| 573 |
+
"step": 270,
|
| 574 |
+
"token_acc": 0.8236703843672228,
|
| 575 |
+
"train_speed(iter/s)": 0.006963
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.535801266439357,
|
| 579 |
+
"grad_norm": 0.4944189786911011,
|
| 580 |
+
"learning_rate": 4.8226372212343726e-05,
|
| 581 |
+
"loss": 0.5667993545532226,
|
| 582 |
+
"memory(GiB)": 291.53,
|
| 583 |
+
"step": 275,
|
| 584 |
+
"token_acc": 0.8146758992113868,
|
| 585 |
+
"train_speed(iter/s)": 0.006964
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.5455431076473454,
|
| 589 |
+
"grad_norm": 0.8255349397659302,
|
| 590 |
+
"learning_rate": 4.661586191227247e-05,
|
| 591 |
+
"loss": 0.5635212898254395,
|
| 592 |
+
"memory(GiB)": 291.53,
|
| 593 |
+
"step": 280,
|
| 594 |
+
"token_acc": 0.8212789502806103,
|
| 595 |
+
"train_speed(iter/s)": 0.006981
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 0.5552849488553336,
|
| 599 |
+
"grad_norm": 0.4205448627471924,
|
| 600 |
+
"learning_rate": 4.500887201417187e-05,
|
| 601 |
+
"loss": 0.6006342887878418,
|
| 602 |
+
"memory(GiB)": 291.53,
|
| 603 |
+
"step": 285,
|
| 604 |
+
"token_acc": 0.8099815913274698,
|
| 605 |
+
"train_speed(iter/s)": 0.007001
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.565026790063322,
|
| 609 |
+
"grad_norm": 0.39303454756736755,
|
| 610 |
+
"learning_rate": 4.3407074214130446e-05,
|
| 611 |
+
"loss": 0.5548898220062256,
|
| 612 |
+
"memory(GiB)": 291.53,
|
| 613 |
+
"step": 290,
|
| 614 |
+
"token_acc": 0.8264817150063052,
|
| 615 |
+
"train_speed(iter/s)": 0.007007
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 0.5747686312713103,
|
| 619 |
+
"grad_norm": 0.5799729824066162,
|
| 620 |
+
"learning_rate": 4.181213480707637e-05,
|
| 621 |
+
"loss": 0.5500263214111328,
|
| 622 |
+
"memory(GiB)": 291.53,
|
| 623 |
+
"step": 295,
|
| 624 |
+
"token_acc": 0.8245773863751392,
|
| 625 |
+
"train_speed(iter/s)": 0.007016
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.5845104724792985,
|
| 629 |
+
"grad_norm": 0.5087621212005615,
|
| 630 |
+
"learning_rate": 4.0225712953388494e-05,
|
| 631 |
+
"loss": 0.5443117618560791,
|
| 632 |
+
"memory(GiB)": 291.53,
|
| 633 |
+
"step": 300,
|
| 634 |
+
"token_acc": 0.8282135401133314,
|
| 635 |
+
"train_speed(iter/s)": 0.007015
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.5845104724792985,
|
| 639 |
+
"eval_loss": 0.5304377675056458,
|
| 640 |
+
"eval_runtime": 26.9276,
|
| 641 |
+
"eval_samples_per_second": 0.149,
|
| 642 |
+
"eval_steps_per_second": 0.074,
|
| 643 |
+
"step": 300
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 0.5942523136872869,
|
| 647 |
+
"grad_norm": 0.5076440572738647,
|
| 648 |
+
"learning_rate": 3.864945895292908e-05,
|
| 649 |
+
"loss": 0.6255448818206787,
|
| 650 |
+
"memory(GiB)": 291.53,
|
| 651 |
+
"step": 305,
|
| 652 |
+
"token_acc": 0.8085856491082538,
|
| 653 |
+
"train_speed(iter/s)": 0.007003
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.6039941548952752,
|
| 657 |
+
"grad_norm": 0.5984292030334473,
|
| 658 |
+
"learning_rate": 3.708501252829386e-05,
|
| 659 |
+
"loss": 0.5672544956207275,
|
| 660 |
+
"memory(GiB)": 291.53,
|
| 661 |
+
"step": 310,
|
| 662 |
+
"token_acc": 0.8221675325859084,
|
| 663 |
+
"train_speed(iter/s)": 0.007004
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 0.6137359961032636,
|
| 667 |
+
"grad_norm": 0.4321894645690918,
|
| 668 |
+
"learning_rate": 3.553400111906523e-05,
|
| 669 |
+
"loss": 0.5947960376739502,
|
| 670 |
+
"memory(GiB)": 291.53,
|
| 671 |
+
"step": 315,
|
| 672 |
+
"token_acc": 0.8143758155719878,
|
| 673 |
+
"train_speed(iter/s)": 0.007012
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.6234778373112518,
|
| 677 |
+
"grad_norm": 0.7790753245353699,
|
| 678 |
+
"learning_rate": 3.399803818884311e-05,
|
| 679 |
+
"loss": 0.5950922489166259,
|
| 680 |
+
"memory(GiB)": 291.53,
|
| 681 |
+
"step": 320,
|
| 682 |
+
"token_acc": 0.8138864880437914,
|
| 683 |
+
"train_speed(iter/s)": 0.007019
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.6332196785192401,
|
| 687 |
+
"grad_norm": 0.37229499220848083,
|
| 688 |
+
"learning_rate": 3.247872154681439e-05,
|
| 689 |
+
"loss": 0.5663117408752442,
|
| 690 |
+
"memory(GiB)": 291.53,
|
| 691 |
+
"step": 325,
|
| 692 |
+
"token_acc": 0.8152720218675602,
|
| 693 |
+
"train_speed(iter/s)": 0.007009
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 0.6429615197272285,
|
| 697 |
+
"grad_norm": 0.5171226859092712,
|
| 698 |
+
"learning_rate": 3.097763168560741e-05,
|
| 699 |
+
"loss": 0.5201399803161622,
|
| 700 |
+
"memory(GiB)": 291.53,
|
| 701 |
+
"step": 330,
|
| 702 |
+
"token_acc": 0.8316269396896496,
|
| 703 |
+
"train_speed(iter/s)": 0.007019
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.6527033609352167,
|
| 707 |
+
"grad_norm": 0.579806923866272,
|
| 708 |
+
"learning_rate": 2.949633013715982e-05,
|
| 709 |
+
"loss": 0.576199722290039,
|
| 710 |
+
"memory(GiB)": 291.53,
|
| 711 |
+
"step": 335,
|
| 712 |
+
"token_acc": 0.8226395939086294,
|
| 713 |
+
"train_speed(iter/s)": 0.007018
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.662445202143205,
|
| 717 |
+
"grad_norm": 0.5151726007461548,
|
| 718 |
+
"learning_rate": 2.8036357848311012e-05,
|
| 719 |
+
"loss": 0.5100949287414551,
|
| 720 |
+
"memory(GiB)": 291.53,
|
| 721 |
+
"step": 340,
|
| 722 |
+
"token_acc": 0.835439610037913,
|
| 723 |
+
"train_speed(iter/s)": 0.007019
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.6721870433511934,
|
| 727 |
+
"grad_norm": 0.6563583612442017,
|
| 728 |
+
"learning_rate": 2.659923357780828e-05,
|
| 729 |
+
"loss": 0.5569330215454101,
|
| 730 |
+
"memory(GiB)": 291.53,
|
| 731 |
+
"step": 345,
|
| 732 |
+
"token_acc": 0.8189111231296197,
|
| 733 |
+
"train_speed(iter/s)": 0.007026
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 0.6819288845591817,
|
| 737 |
+
"grad_norm": 0.4978316128253937,
|
| 738 |
+
"learning_rate": 2.518645231639457e-05,
|
| 739 |
+
"loss": 0.5415875434875488,
|
| 740 |
+
"memory(GiB)": 291.53,
|
| 741 |
+
"step": 350,
|
| 742 |
+
"token_acc": 0.8224528817263043,
|
| 743 |
+
"train_speed(iter/s)": 0.007018
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.69167072576717,
|
| 747 |
+
"grad_norm": 0.3655760586261749,
|
| 748 |
+
"learning_rate": 2.3799483731621237e-05,
|
| 749 |
+
"loss": 0.5323990821838379,
|
| 750 |
+
"memory(GiB)": 291.53,
|
| 751 |
+
"step": 355,
|
| 752 |
+
"token_acc": 0.8286348880915769,
|
| 753 |
+
"train_speed(iter/s)": 0.007019
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.7014125669751583,
|
| 757 |
+
"grad_norm": 0.5881947875022888,
|
| 758 |
+
"learning_rate": 2.2439770639003627e-05,
|
| 759 |
+
"loss": 0.5367710590362549,
|
| 760 |
+
"memory(GiB)": 291.53,
|
| 761 |
+
"step": 360,
|
| 762 |
+
"token_acc": 0.8259178446067735,
|
| 763 |
+
"train_speed(iter/s)": 0.007019
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 0.7111544081831466,
|
| 767 |
+
"grad_norm": 0.4462352991104126,
|
| 768 |
+
"learning_rate": 2.110872750110996e-05,
|
| 769 |
+
"loss": 0.6012316703796386,
|
| 770 |
+
"memory(GiB)": 291.53,
|
| 771 |
+
"step": 365,
|
| 772 |
+
"token_acc": 0.8113236655353286,
|
| 773 |
+
"train_speed(iter/s)": 0.007025
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.720896249391135,
|
| 777 |
+
"grad_norm": 0.5265180468559265,
|
| 778 |
+
"learning_rate": 1.980773895614481e-05,
|
| 779 |
+
"loss": 0.5925416946411133,
|
| 780 |
+
"memory(GiB)": 291.53,
|
| 781 |
+
"step": 370,
|
| 782 |
+
"token_acc": 0.8087358252834943,
|
| 783 |
+
"train_speed(iter/s)": 0.007037
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 0.7306380905991232,
|
| 787 |
+
"grad_norm": 0.5589653253555298,
|
| 788 |
+
"learning_rate": 1.8538158377557702e-05,
|
| 789 |
+
"loss": 0.554245138168335,
|
| 790 |
+
"memory(GiB)": 291.53,
|
| 791 |
+
"step": 375,
|
| 792 |
+
"token_acc": 0.8206760127289928,
|
| 793 |
+
"train_speed(iter/s)": 0.007042
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.7403799318071116,
|
| 797 |
+
"grad_norm": 0.49020129442214966,
|
| 798 |
+
"learning_rate": 1.7301306466175533e-05,
|
| 799 |
+
"loss": 0.5485538482666016,
|
| 800 |
+
"memory(GiB)": 291.53,
|
| 801 |
+
"step": 380,
|
| 802 |
+
"token_acc": 0.8174893357708715,
|
| 803 |
+
"train_speed(iter/s)": 0.007043
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.7501217730150999,
|
| 807 |
+
"grad_norm": 0.8308594226837158,
|
| 808 |
+
"learning_rate": 1.6098469876323093e-05,
|
| 809 |
+
"loss": 0.5580153465270996,
|
| 810 |
+
"memory(GiB)": 291.53,
|
| 811 |
+
"step": 385,
|
| 812 |
+
"token_acc": 0.8252497729336966,
|
| 813 |
+
"train_speed(iter/s)": 0.007044
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.7598636142230881,
|
| 817 |
+
"grad_norm": 0.9425000548362732,
|
| 818 |
+
"learning_rate": 1.4930899877361015e-05,
|
| 819 |
+
"loss": 0.5678109169006348,
|
| 820 |
+
"memory(GiB)": 291.53,
|
| 821 |
+
"step": 390,
|
| 822 |
+
"token_acc": 0.8189367271414387,
|
| 823 |
+
"train_speed(iter/s)": 0.007047
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.7696054554310765,
|
| 827 |
+
"grad_norm": 0.46440383791923523,
|
| 828 |
+
"learning_rate": 1.3799811052033467e-05,
|
| 829 |
+
"loss": 0.5055109024047851,
|
| 830 |
+
"memory(GiB)": 291.53,
|
| 831 |
+
"step": 395,
|
| 832 |
+
"token_acc": 0.835208145167912,
|
| 833 |
+
"train_speed(iter/s)": 0.007044
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.7793472966390648,
|
| 837 |
+
"grad_norm": 0.6887357831001282,
|
| 838 |
+
"learning_rate": 1.2706380032979691e-05,
|
| 839 |
+
"loss": 0.6191585540771485,
|
| 840 |
+
"memory(GiB)": 291.53,
|
| 841 |
+
"step": 400,
|
| 842 |
+
"token_acc": 0.801925584194163,
|
| 843 |
+
"train_speed(iter/s)": 0.007049
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.7793472966390648,
|
| 847 |
+
"eval_loss": 0.5201972723007202,
|
| 848 |
+
"eval_runtime": 26.7707,
|
| 849 |
+
"eval_samples_per_second": 0.149,
|
| 850 |
+
"eval_steps_per_second": 0.075,
|
| 851 |
+
"step": 400
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.7890891378470531,
|
| 855 |
+
"grad_norm": 0.6512044072151184,
|
| 856 |
+
"learning_rate": 1.1651744278723687e-05,
|
| 857 |
+
"loss": 0.5115838527679444,
|
| 858 |
+
"memory(GiB)": 291.53,
|
| 859 |
+
"step": 405,
|
| 860 |
+
"token_acc": 0.8362541073384447,
|
| 861 |
+
"train_speed(iter/s)": 0.007043
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 0.7988309790550414,
|
| 865 |
+
"grad_norm": 0.36380794644355774,
|
| 866 |
+
"learning_rate": 1.0637000890415388e-05,
|
| 867 |
+
"loss": 0.5678341865539551,
|
| 868 |
+
"memory(GiB)": 291.53,
|
| 869 |
+
"step": 410,
|
| 870 |
+
"token_acc": 0.8218965517241379,
|
| 871 |
+
"train_speed(iter/s)": 0.007056
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.8085728202630297,
|
| 875 |
+
"grad_norm": 0.524336576461792,
|
| 876 |
+
"learning_rate": 9.663205470554276e-06,
|
| 877 |
+
"loss": 0.5683213233947754,
|
| 878 |
+
"memory(GiB)": 291.54,
|
| 879 |
+
"step": 415,
|
| 880 |
+
"token_acc": 0.8144156614593633,
|
| 881 |
+
"train_speed(iter/s)": 0.007057
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.8183146614710181,
|
| 885 |
+
"grad_norm": 0.510553777217865,
|
| 886 |
+
"learning_rate": 8.73137102488249e-06,
|
| 887 |
+
"loss": 0.5425020217895508,
|
| 888 |
+
"memory(GiB)": 291.54,
|
| 889 |
+
"step": 420,
|
| 890 |
+
"token_acc": 0.8262775083923909,
|
| 891 |
+
"train_speed(iter/s)": 0.007067
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.8280565026790063,
|
| 895 |
+
"grad_norm": 0.4342099726200104,
|
| 896 |
+
"learning_rate": 7.842466908590006e-06,
|
| 897 |
+
"loss": 0.4687312602996826,
|
| 898 |
+
"memory(GiB)": 291.54,
|
| 899 |
+
"step": 425,
|
| 900 |
+
"token_acc": 0.8489349775784754,
|
| 901 |
+
"train_speed(iter/s)": 0.007065
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.8377983438869946,
|
| 905 |
+
"grad_norm": 0.47141674160957336,
|
| 906 |
+
"learning_rate": 6.997417817927865e-06,
|
| 907 |
+
"loss": 0.5220559597015381,
|
| 908 |
+
"memory(GiB)": 291.54,
|
| 909 |
+
"step": 430,
|
| 910 |
+
"token_acc": 0.8345159448658348,
|
| 911 |
+
"train_speed(iter/s)": 0.007065
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.847540185094983,
|
| 915 |
+
"grad_norm": 0.7017420530319214,
|
| 916 |
+
"learning_rate": 6.197102828278611e-06,
|
| 917 |
+
"loss": 0.4900541305541992,
|
| 918 |
+
"memory(GiB)": 291.54,
|
| 919 |
+
"step": 435,
|
| 920 |
+
"token_acc": 0.846220772158782,
|
| 921 |
+
"train_speed(iter/s)": 0.007061
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.8572820263029712,
|
| 925 |
+
"grad_norm": 0.9358065724372864,
|
| 926 |
+
"learning_rate": 5.442354479684558e-06,
|
| 927 |
+
"loss": 0.5937717914581299,
|
| 928 |
+
"memory(GiB)": 291.54,
|
| 929 |
+
"step": 440,
|
| 930 |
+
"token_acc": 0.8115085230071928,
|
| 931 |
+
"train_speed(iter/s)": 0.007068
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 0.8670238675109596,
|
| 935 |
+
"grad_norm": 0.6291302442550659,
|
| 936 |
+
"learning_rate": 4.733957910785114e-06,
|
| 937 |
+
"loss": 0.5440679073333741,
|
| 938 |
+
"memory(GiB)": 291.54,
|
| 939 |
+
"step": 445,
|
| 940 |
+
"token_acc": 0.8238650458943191,
|
| 941 |
+
"train_speed(iter/s)": 0.007071
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.8767657087189479,
|
| 945 |
+
"grad_norm": 0.7625430822372437,
|
| 946 |
+
"learning_rate": 4.072650042064174e-06,
|
| 947 |
+
"loss": 0.5928035736083984,
|
| 948 |
+
"memory(GiB)": 291.54,
|
| 949 |
+
"step": 450,
|
| 950 |
+
"token_acc": 0.8072507552870091,
|
| 951 |
+
"train_speed(iter/s)": 0.007064
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 0.8865075499269361,
|
| 955 |
+
"grad_norm": 0.5580248832702637,
|
| 956 |
+
"learning_rate": 3.4591188092571893e-06,
|
| 957 |
+
"loss": 0.5281401634216308,
|
| 958 |
+
"memory(GiB)": 291.54,
|
| 959 |
+
"step": 455,
|
| 960 |
+
"token_acc": 0.8284527518172378,
|
| 961 |
+
"train_speed(iter/s)": 0.007059
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.8962493911349245,
|
| 965 |
+
"grad_norm": 0.6311898827552795,
|
| 966 |
+
"learning_rate": 2.894002447715399e-06,
|
| 967 |
+
"loss": 0.483402156829834,
|
| 968 |
+
"memory(GiB)": 291.54,
|
| 969 |
+
"step": 460,
|
| 970 |
+
"token_acc": 0.846503178928247,
|
| 971 |
+
"train_speed(iter/s)": 0.007059
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 0.9059912323429128,
|
| 975 |
+
"grad_norm": 0.47069641947746277,
|
| 976 |
+
"learning_rate": 2.3778888284716193e-06,
|
| 977 |
+
"loss": 0.5357548236846924,
|
| 978 |
+
"memory(GiB)": 291.54,
|
| 979 |
+
"step": 465,
|
| 980 |
+
"token_acc": 0.8302510497632449,
|
| 981 |
+
"train_speed(iter/s)": 0.007062
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.9157330735509012,
|
| 985 |
+
"grad_norm": 0.6504586935043335,
|
| 986 |
+
"learning_rate": 1.9113148466983254e-06,
|
| 987 |
+
"loss": 0.4945201873779297,
|
| 988 |
+
"memory(GiB)": 291.54,
|
| 989 |
+
"step": 470,
|
| 990 |
+
"token_acc": 0.836600057366861,
|
| 991 |
+
"train_speed(iter/s)": 0.007062
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.9254749147588894,
|
| 995 |
+
"grad_norm": 0.8726986646652222,
|
| 996 |
+
"learning_rate": 1.4947658631941309e-06,
|
| 997 |
+
"loss": 0.6114956855773925,
|
| 998 |
+
"memory(GiB)": 291.54,
|
| 999 |
+
"step": 475,
|
| 1000 |
+
"token_acc": 0.8079200232941861,
|
| 1001 |
+
"train_speed(iter/s)": 0.007069
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.9352167559668777,
|
| 1005 |
+
"grad_norm": 0.4940550625324249,
|
| 1006 |
+
"learning_rate": 1.1286751994797284e-06,
|
| 1007 |
+
"loss": 0.4838570117950439,
|
| 1008 |
+
"memory(GiB)": 291.54,
|
| 1009 |
+
"step": 480,
|
| 1010 |
+
"token_acc": 0.8428108691740147,
|
| 1011 |
+
"train_speed(iter/s)": 0.007071
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.9449585971748661,
|
| 1015 |
+
"grad_norm": 0.47085583209991455,
|
| 1016 |
+
"learning_rate": 8.134236870284861e-07,
|
| 1017 |
+
"loss": 0.5545851230621338,
|
| 1018 |
+
"memory(GiB)": 291.54,
|
| 1019 |
+
"step": 485,
|
| 1020 |
+
"token_acc": 0.8195587462791105,
|
| 1021 |
+
"train_speed(iter/s)": 0.007076
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.9547004383828543,
|
| 1025 |
+
"grad_norm": 0.47154033184051514,
|
| 1026 |
+
"learning_rate": 5.493392711005796e-07,
|
| 1027 |
+
"loss": 0.5688926696777343,
|
| 1028 |
+
"memory(GiB)": 291.54,
|
| 1029 |
+
"step": 490,
|
| 1030 |
+
"token_acc": 0.8178977272727272,
|
| 1031 |
+
"train_speed(iter/s)": 0.007077
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.9644422795908427,
|
| 1035 |
+
"grad_norm": 0.49407562613487244,
|
| 1036 |
+
"learning_rate": 3.366966695929119e-07,
|
| 1037 |
+
"loss": 0.5036429405212403,
|
| 1038 |
+
"memory(GiB)": 291.54,
|
| 1039 |
+
"step": 495,
|
| 1040 |
+
"token_acc": 0.8390853451207072,
|
| 1041 |
+
"train_speed(iter/s)": 0.007075
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 0.974184120798831,
|
| 1045 |
+
"grad_norm": 0.5743114352226257,
|
| 1046 |
+
"learning_rate": 1.7571708725953596e-07,
|
| 1047 |
+
"loss": 0.514036750793457,
|
| 1048 |
+
"memory(GiB)": 291.54,
|
| 1049 |
+
"step": 500,
|
| 1050 |
+
"token_acc": 0.8316027940343591,
|
| 1051 |
+
"train_speed(iter/s)": 0.007081
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.974184120798831,
|
| 1055 |
+
"eval_loss": 0.5129119753837585,
|
| 1056 |
+
"eval_runtime": 26.8989,
|
| 1057 |
+
"eval_samples_per_second": 0.149,
|
| 1058 |
+
"eval_steps_per_second": 0.074,
|
| 1059 |
+
"step": 500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.9839259620068193,
|
| 1063 |
+
"grad_norm": 0.5348408818244934,
|
| 1064 |
+
"learning_rate": 6.656798560001343e-08,
|
| 1065 |
+
"loss": 0.5319667816162109,
|
| 1066 |
+
"memory(GiB)": 291.54,
|
| 1067 |
+
"step": 505,
|
| 1068 |
+
"token_acc": 0.8256676557863502,
|
| 1069 |
+
"train_speed(iter/s)": 0.007071
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 0.9936678032148076,
|
| 1073 |
+
"grad_norm": 0.5291311144828796,
|
| 1074 |
+
"learning_rate": 9.362908654986235e-09,
|
| 1075 |
+
"loss": 0.5844589233398437,
|
| 1076 |
+
"memory(GiB)": 291.54,
|
| 1077 |
+
"step": 510,
|
| 1078 |
+
"token_acc": 0.8063741960257271,
|
| 1079 |
+
"train_speed(iter/s)": 0.007071
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.9995129079396006,
|
| 1083 |
+
"eval_loss": 0.5115394592285156,
|
| 1084 |
+
"eval_runtime": 26.9687,
|
| 1085 |
+
"eval_samples_per_second": 0.148,
|
| 1086 |
+
"eval_steps_per_second": 0.074,
|
| 1087 |
+
"step": 513
|
| 1088 |
+
}
|
| 1089 |
+
],
|
| 1090 |
+
"logging_steps": 5,
|
| 1091 |
+
"max_steps": 513,
|
| 1092 |
+
"num_input_tokens_seen": 0,
|
| 1093 |
+
"num_train_epochs": 1,
|
| 1094 |
+
"save_steps": 50,
|
| 1095 |
+
"stateful_callbacks": {
|
| 1096 |
+
"TrainerControl": {
|
| 1097 |
+
"args": {
|
| 1098 |
+
"should_epoch_stop": false,
|
| 1099 |
+
"should_evaluate": false,
|
| 1100 |
+
"should_log": false,
|
| 1101 |
+
"should_save": true,
|
| 1102 |
+
"should_training_stop": true
|
| 1103 |
+
},
|
| 1104 |
+
"attributes": {}
|
| 1105 |
+
}
|
| 1106 |
+
},
|
| 1107 |
+
"total_flos": 8.13335087128363e+18,
|
| 1108 |
+
"train_batch_size": 2,
|
| 1109 |
+
"trial_name": null,
|
| 1110 |
+
"trial_params": null
|
| 1111 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8dec67b291f512f8ffdff21456d56f0fecdcf895721c18f5edb1e591680d3bd0
|
| 3 |
+
size 5816
|