Commit
•
3be5e0e
1
Parent(s):
8182378
Model save
Browse files- trainer_state.json +64 -12
- training_args.bin +1 -1
trainer_state.json
CHANGED
|
@@ -1,28 +1,80 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"eval_steps":
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 3.0,
|
| 13 |
-
"
|
| 14 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
"train_loss": 0.0,
|
| 16 |
-
"train_runtime":
|
| 17 |
-
"train_samples_per_second":
|
| 18 |
-
"train_steps_per_second":
|
| 19 |
}
|
| 20 |
],
|
| 21 |
-
"logging_steps":
|
| 22 |
-
"max_steps":
|
| 23 |
-
"num_train_epochs":
|
| 24 |
"save_steps": 500,
|
| 25 |
-
"total_flos":
|
| 26 |
"trial_name": null,
|
| 27 |
"trial_params": null
|
| 28 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 4,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_gen_len": 18.4,
|
| 14 |
+
"eval_loss": NaN,
|
| 15 |
+
"eval_rouge1": 0.04,
|
| 16 |
+
"eval_rouge2": 0.0,
|
| 17 |
+
"eval_rougeL": 0.04,
|
| 18 |
+
"eval_rougeLsum": 0.04,
|
| 19 |
+
"eval_runtime": 1.5114,
|
| 20 |
+
"eval_samples_per_second": 6.616,
|
| 21 |
+
"eval_steps_per_second": 0.662,
|
| 22 |
+
"step": 1
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 2.0,
|
| 26 |
+
"eval_gen_len": 18.4,
|
| 27 |
+
"eval_loss": NaN,
|
| 28 |
+
"eval_rouge1": 0.04,
|
| 29 |
+
"eval_rouge2": 0.0,
|
| 30 |
+
"eval_rougeL": 0.04,
|
| 31 |
+
"eval_rougeLsum": 0.04,
|
| 32 |
+
"eval_runtime": 0.5228,
|
| 33 |
+
"eval_samples_per_second": 19.127,
|
| 34 |
+
"eval_steps_per_second": 1.913,
|
| 35 |
+
"step": 2
|
| 36 |
+
},
|
| 37 |
{
|
| 38 |
"epoch": 3.0,
|
| 39 |
+
"eval_gen_len": 18.4,
|
| 40 |
+
"eval_loss": NaN,
|
| 41 |
+
"eval_rouge1": 0.04,
|
| 42 |
+
"eval_rouge2": 0.0,
|
| 43 |
+
"eval_rougeL": 0.04,
|
| 44 |
+
"eval_rougeLsum": 0.04,
|
| 45 |
+
"eval_runtime": 0.5146,
|
| 46 |
+
"eval_samples_per_second": 19.434,
|
| 47 |
+
"eval_steps_per_second": 1.943,
|
| 48 |
+
"step": 3
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 4.0,
|
| 52 |
+
"eval_gen_len": 18.4,
|
| 53 |
+
"eval_loss": NaN,
|
| 54 |
+
"eval_rouge1": 0.04,
|
| 55 |
+
"eval_rouge2": 0.0,
|
| 56 |
+
"eval_rougeL": 0.04,
|
| 57 |
+
"eval_rougeLsum": 0.04,
|
| 58 |
+
"eval_runtime": 0.5111,
|
| 59 |
+
"eval_samples_per_second": 19.567,
|
| 60 |
+
"eval_steps_per_second": 1.957,
|
| 61 |
+
"step": 4
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 4.0,
|
| 65 |
+
"step": 4,
|
| 66 |
+
"total_flos": 7435620188160.0,
|
| 67 |
"train_loss": 0.0,
|
| 68 |
+
"train_runtime": 3.7792,
|
| 69 |
+
"train_samples_per_second": 10.584,
|
| 70 |
+
"train_steps_per_second": 1.058
|
| 71 |
}
|
| 72 |
],
|
| 73 |
+
"logging_steps": 500,
|
| 74 |
+
"max_steps": 4,
|
| 75 |
+
"num_train_epochs": 4,
|
| 76 |
"save_steps": 500,
|
| 77 |
+
"total_flos": 7435620188160.0,
|
| 78 |
"trial_name": null,
|
| 79 |
"trial_params": null
|
| 80 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4155
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c3058916ffeb55eb6dddcb4252c3c0b496033bbd44e5e22a2047f428bc39317
|
| 3 |
size 4155
|