End of training
Browse files- README.md +14 -2
- all_results.json +18 -0
- eval_results.json +13 -0
- train_results.json +8 -0
- trainer_state.json +960 -0
README.md
CHANGED
|
@@ -3,11 +3,23 @@ license: apache-2.0
|
|
| 3 |
base_model: google/long-t5-tglobal-xl
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
metrics:
|
| 7 |
- rouge
|
| 8 |
model-index:
|
| 9 |
- name: longt5_xl_sfd_bp_15
|
| 10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -15,7 +27,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# longt5_xl_sfd_bp_15
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [google/long-t5-tglobal-xl](https://huggingface.co/google/long-t5-tglobal-xl) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 2.5840
|
| 21 |
- Rouge1: 29.7482
|
|
|
|
| 3 |
base_model: google/long-t5-tglobal-xl
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- learn3r/summ_screen_fd_bp
|
| 8 |
metrics:
|
| 9 |
- rouge
|
| 10 |
model-index:
|
| 11 |
- name: longt5_xl_sfd_bp_15
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Summarization
|
| 15 |
+
type: summarization
|
| 16 |
+
dataset:
|
| 17 |
+
name: learn3r/summ_screen_fd_bp
|
| 18 |
+
type: learn3r/summ_screen_fd_bp
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Rouge1
|
| 21 |
+
type: rouge
|
| 22 |
+
value: 29.7482
|
| 23 |
---
|
| 24 |
|
| 25 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 27 |
|
| 28 |
# longt5_xl_sfd_bp_15
|
| 29 |
|
| 30 |
+
This model is a fine-tuned version of [google/long-t5-tglobal-xl](https://huggingface.co/google/long-t5-tglobal-xl) on the learn3r/summ_screen_fd_bp dataset.
|
| 31 |
It achieves the following results on the evaluation set:
|
| 32 |
- Loss: 2.5840
|
| 33 |
- Rouge1: 29.7482
|
all_results.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.61,
|
| 3 |
+
"eval_gen_len": 503.5769230769231,
|
| 4 |
+
"eval_loss": 2.5840089321136475,
|
| 5 |
+
"eval_rouge1": 29.7482,
|
| 6 |
+
"eval_rouge2": 12.0072,
|
| 7 |
+
"eval_rougeL": 21.348,
|
| 8 |
+
"eval_rougeLsum": 28.5849,
|
| 9 |
+
"eval_runtime": 1799.7554,
|
| 10 |
+
"eval_samples": 338,
|
| 11 |
+
"eval_samples_per_second": 0.188,
|
| 12 |
+
"eval_steps_per_second": 0.024,
|
| 13 |
+
"train_loss": 0.9784720075981957,
|
| 14 |
+
"train_runtime": 78497.3694,
|
| 15 |
+
"train_samples": 3673,
|
| 16 |
+
"train_samples_per_second": 0.702,
|
| 17 |
+
"train_steps_per_second": 0.003
|
| 18 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.61,
|
| 3 |
+
"eval_gen_len": 503.5769230769231,
|
| 4 |
+
"eval_loss": 2.5840089321136475,
|
| 5 |
+
"eval_rouge1": 29.7482,
|
| 6 |
+
"eval_rouge2": 12.0072,
|
| 7 |
+
"eval_rougeL": 21.348,
|
| 8 |
+
"eval_rougeLsum": 28.5849,
|
| 9 |
+
"eval_runtime": 1799.7554,
|
| 10 |
+
"eval_samples": 338,
|
| 11 |
+
"eval_samples_per_second": 0.188,
|
| 12 |
+
"eval_steps_per_second": 0.024
|
| 13 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.61,
|
| 3 |
+
"train_loss": 0.9784720075981957,
|
| 4 |
+
"train_runtime": 78497.3694,
|
| 5 |
+
"train_samples": 3673,
|
| 6 |
+
"train_samples_per_second": 0.702,
|
| 7 |
+
"train_steps_per_second": 0.003
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,960 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 25.2438,
|
| 3 |
+
"best_model_checkpoint": "/exports/eddie/scratch/s1970716/models/longt5_xl_sfd_bp_15/checkpoint-201",
|
| 4 |
+
"epoch": 14.608695652173914,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 210,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.14,
|
| 13 |
+
"grad_norm": 4.218067169189453,
|
| 14 |
+
"learning_rate": 0.001,
|
| 15 |
+
"loss": 2.9668,
|
| 16 |
+
"step": 2
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.28,
|
| 20 |
+
"grad_norm": 200.38365173339844,
|
| 21 |
+
"learning_rate": 0.001,
|
| 22 |
+
"loss": 3.3198,
|
| 23 |
+
"step": 4
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.42,
|
| 27 |
+
"grad_norm": 74.55081176757812,
|
| 28 |
+
"learning_rate": 0.001,
|
| 29 |
+
"loss": 2.6874,
|
| 30 |
+
"step": 6
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.56,
|
| 34 |
+
"grad_norm": 19.85554313659668,
|
| 35 |
+
"learning_rate": 0.001,
|
| 36 |
+
"loss": 2.3138,
|
| 37 |
+
"step": 8
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.7,
|
| 41 |
+
"grad_norm": 41.041751861572266,
|
| 42 |
+
"learning_rate": 0.001,
|
| 43 |
+
"loss": 2.0222,
|
| 44 |
+
"step": 10
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.83,
|
| 48 |
+
"grad_norm": 3.789278507232666,
|
| 49 |
+
"learning_rate": 0.001,
|
| 50 |
+
"loss": 1.8363,
|
| 51 |
+
"step": 12
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.97,
|
| 55 |
+
"grad_norm": 5.874878883361816,
|
| 56 |
+
"learning_rate": 0.001,
|
| 57 |
+
"loss": 2.5763,
|
| 58 |
+
"step": 14
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.97,
|
| 62 |
+
"eval_gen_len": 509.64792899408286,
|
| 63 |
+
"eval_loss": 2.541541814804077,
|
| 64 |
+
"eval_rouge1": 10.6052,
|
| 65 |
+
"eval_rouge2": 1.4494,
|
| 66 |
+
"eval_rougeL": 10.4593,
|
| 67 |
+
"eval_rougeLsum": 10.4801,
|
| 68 |
+
"eval_runtime": 1798.047,
|
| 69 |
+
"eval_samples_per_second": 0.188,
|
| 70 |
+
"eval_steps_per_second": 0.024,
|
| 71 |
+
"step": 14
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.11,
|
| 75 |
+
"grad_norm": 1.338165521621704,
|
| 76 |
+
"learning_rate": 0.001,
|
| 77 |
+
"loss": 2.4441,
|
| 78 |
+
"step": 16
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.25,
|
| 82 |
+
"grad_norm": 3.755629777908325,
|
| 83 |
+
"learning_rate": 0.001,
|
| 84 |
+
"loss": 2.258,
|
| 85 |
+
"step": 18
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 1.39,
|
| 89 |
+
"grad_norm": 6.490938663482666,
|
| 90 |
+
"learning_rate": 0.001,
|
| 91 |
+
"loss": 3.0147,
|
| 92 |
+
"step": 20
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.53,
|
| 96 |
+
"grad_norm": 5.593593597412109,
|
| 97 |
+
"learning_rate": 0.001,
|
| 98 |
+
"loss": 2.4724,
|
| 99 |
+
"step": 22
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.67,
|
| 103 |
+
"grad_norm": 1.0521235466003418,
|
| 104 |
+
"learning_rate": 0.001,
|
| 105 |
+
"loss": 2.023,
|
| 106 |
+
"step": 24
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.81,
|
| 110 |
+
"grad_norm": 12.585270881652832,
|
| 111 |
+
"learning_rate": 0.001,
|
| 112 |
+
"loss": 2.223,
|
| 113 |
+
"step": 26
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.95,
|
| 117 |
+
"grad_norm": 78.1630630493164,
|
| 118 |
+
"learning_rate": 0.001,
|
| 119 |
+
"loss": 1.8998,
|
| 120 |
+
"step": 28
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.95,
|
| 124 |
+
"eval_gen_len": 511.0,
|
| 125 |
+
"eval_loss": 1.739753246307373,
|
| 126 |
+
"eval_rouge1": 16.7989,
|
| 127 |
+
"eval_rouge2": 4.1457,
|
| 128 |
+
"eval_rougeL": 16.4049,
|
| 129 |
+
"eval_rougeLsum": 15.1803,
|
| 130 |
+
"eval_runtime": 1798.9905,
|
| 131 |
+
"eval_samples_per_second": 0.188,
|
| 132 |
+
"eval_steps_per_second": 0.024,
|
| 133 |
+
"step": 28
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 2.09,
|
| 137 |
+
"grad_norm": 0.7915446758270264,
|
| 138 |
+
"learning_rate": 0.001,
|
| 139 |
+
"loss": 1.8375,
|
| 140 |
+
"step": 30
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 2.23,
|
| 144 |
+
"grad_norm": 1.5996413230895996,
|
| 145 |
+
"learning_rate": 0.001,
|
| 146 |
+
"loss": 2.0326,
|
| 147 |
+
"step": 32
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.37,
|
| 151 |
+
"grad_norm": 1.0431970357894897,
|
| 152 |
+
"learning_rate": 0.001,
|
| 153 |
+
"loss": 2.1242,
|
| 154 |
+
"step": 34
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 2.5,
|
| 158 |
+
"grad_norm": 0.5979584455490112,
|
| 159 |
+
"learning_rate": 0.001,
|
| 160 |
+
"loss": 2.0047,
|
| 161 |
+
"step": 36
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 2.64,
|
| 165 |
+
"grad_norm": 0.28407618403434753,
|
| 166 |
+
"learning_rate": 0.001,
|
| 167 |
+
"loss": 1.7317,
|
| 168 |
+
"step": 38
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 2.78,
|
| 172 |
+
"grad_norm": 0.3217169940471649,
|
| 173 |
+
"learning_rate": 0.001,
|
| 174 |
+
"loss": 1.694,
|
| 175 |
+
"step": 40
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 2.92,
|
| 179 |
+
"grad_norm": 0.33735284209251404,
|
| 180 |
+
"learning_rate": 0.001,
|
| 181 |
+
"loss": 1.6403,
|
| 182 |
+
"step": 42
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 2.99,
|
| 186 |
+
"eval_gen_len": 511.0,
|
| 187 |
+
"eval_loss": 1.5456656217575073,
|
| 188 |
+
"eval_rouge1": 18.4716,
|
| 189 |
+
"eval_rouge2": 5.4633,
|
| 190 |
+
"eval_rougeL": 17.1393,
|
| 191 |
+
"eval_rougeLsum": 16.9242,
|
| 192 |
+
"eval_runtime": 1798.0277,
|
| 193 |
+
"eval_samples_per_second": 0.188,
|
| 194 |
+
"eval_steps_per_second": 0.024,
|
| 195 |
+
"step": 43
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.06,
|
| 199 |
+
"grad_norm": 0.20745837688446045,
|
| 200 |
+
"learning_rate": 0.001,
|
| 201 |
+
"loss": 1.5256,
|
| 202 |
+
"step": 44
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 3.2,
|
| 206 |
+
"grad_norm": 0.24278272688388824,
|
| 207 |
+
"learning_rate": 0.001,
|
| 208 |
+
"loss": 1.4077,
|
| 209 |
+
"step": 46
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 3.34,
|
| 213 |
+
"grad_norm": 3.5210845470428467,
|
| 214 |
+
"learning_rate": 0.001,
|
| 215 |
+
"loss": 1.4244,
|
| 216 |
+
"step": 48
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 3.48,
|
| 220 |
+
"grad_norm": 0.31759026646614075,
|
| 221 |
+
"learning_rate": 0.001,
|
| 222 |
+
"loss": 1.3542,
|
| 223 |
+
"step": 50
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 3.62,
|
| 227 |
+
"grad_norm": 0.2855791449546814,
|
| 228 |
+
"learning_rate": 0.001,
|
| 229 |
+
"loss": 1.3873,
|
| 230 |
+
"step": 52
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 3.76,
|
| 234 |
+
"grad_norm": 0.30171895027160645,
|
| 235 |
+
"learning_rate": 0.001,
|
| 236 |
+
"loss": 1.4693,
|
| 237 |
+
"step": 54
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 3.9,
|
| 241 |
+
"grad_norm": 0.28778406977653503,
|
| 242 |
+
"learning_rate": 0.001,
|
| 243 |
+
"loss": 1.5012,
|
| 244 |
+
"step": 56
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 3.97,
|
| 248 |
+
"eval_gen_len": 511.0,
|
| 249 |
+
"eval_loss": 1.5736442804336548,
|
| 250 |
+
"eval_rouge1": 18.2259,
|
| 251 |
+
"eval_rouge2": 5.3524,
|
| 252 |
+
"eval_rougeL": 17.0162,
|
| 253 |
+
"eval_rougeLsum": 16.7948,
|
| 254 |
+
"eval_runtime": 1799.9735,
|
| 255 |
+
"eval_samples_per_second": 0.188,
|
| 256 |
+
"eval_steps_per_second": 0.024,
|
| 257 |
+
"step": 57
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 4.03,
|
| 261 |
+
"grad_norm": 0.27410924434661865,
|
| 262 |
+
"learning_rate": 0.001,
|
| 263 |
+
"loss": 1.3865,
|
| 264 |
+
"step": 58
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 4.17,
|
| 268 |
+
"grad_norm": 0.2398337870836258,
|
| 269 |
+
"learning_rate": 0.001,
|
| 270 |
+
"loss": 1.198,
|
| 271 |
+
"step": 60
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 4.31,
|
| 275 |
+
"grad_norm": 0.24380528926849365,
|
| 276 |
+
"learning_rate": 0.001,
|
| 277 |
+
"loss": 1.1965,
|
| 278 |
+
"step": 62
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 4.45,
|
| 282 |
+
"grad_norm": 0.28130125999450684,
|
| 283 |
+
"learning_rate": 0.001,
|
| 284 |
+
"loss": 1.2576,
|
| 285 |
+
"step": 64
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.59,
|
| 289 |
+
"grad_norm": 0.22549273073673248,
|
| 290 |
+
"learning_rate": 0.001,
|
| 291 |
+
"loss": 1.2108,
|
| 292 |
+
"step": 66
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 4.73,
|
| 296 |
+
"grad_norm": 0.3336837589740753,
|
| 297 |
+
"learning_rate": 0.001,
|
| 298 |
+
"loss": 1.23,
|
| 299 |
+
"step": 68
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 4.87,
|
| 303 |
+
"grad_norm": 0.39294493198394775,
|
| 304 |
+
"learning_rate": 0.001,
|
| 305 |
+
"loss": 1.248,
|
| 306 |
+
"step": 70
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 4.94,
|
| 310 |
+
"eval_gen_len": 511.0,
|
| 311 |
+
"eval_loss": 1.5482468605041504,
|
| 312 |
+
"eval_rouge1": 20.8275,
|
| 313 |
+
"eval_rouge2": 6.7412,
|
| 314 |
+
"eval_rougeL": 18.0859,
|
| 315 |
+
"eval_rougeLsum": 19.3113,
|
| 316 |
+
"eval_runtime": 1798.5715,
|
| 317 |
+
"eval_samples_per_second": 0.188,
|
| 318 |
+
"eval_steps_per_second": 0.024,
|
| 319 |
+
"step": 71
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 5.01,
|
| 323 |
+
"grad_norm": 0.3731052577495575,
|
| 324 |
+
"learning_rate": 0.001,
|
| 325 |
+
"loss": 1.2523,
|
| 326 |
+
"step": 72
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 5.15,
|
| 330 |
+
"grad_norm": 0.33552882075309753,
|
| 331 |
+
"learning_rate": 0.001,
|
| 332 |
+
"loss": 1.0577,
|
| 333 |
+
"step": 74
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 5.29,
|
| 337 |
+
"grad_norm": 0.3163793087005615,
|
| 338 |
+
"learning_rate": 0.001,
|
| 339 |
+
"loss": 1.0478,
|
| 340 |
+
"step": 76
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 5.43,
|
| 344 |
+
"grad_norm": 0.21926109492778778,
|
| 345 |
+
"learning_rate": 0.001,
|
| 346 |
+
"loss": 1.0127,
|
| 347 |
+
"step": 78
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 5.57,
|
| 351 |
+
"grad_norm": 0.24710944294929504,
|
| 352 |
+
"learning_rate": 0.001,
|
| 353 |
+
"loss": 1.0042,
|
| 354 |
+
"step": 80
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 5.7,
|
| 358 |
+
"grad_norm": 0.2397957742214203,
|
| 359 |
+
"learning_rate": 0.001,
|
| 360 |
+
"loss": 1.0332,
|
| 361 |
+
"step": 82
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 5.84,
|
| 365 |
+
"grad_norm": 0.21428123116493225,
|
| 366 |
+
"learning_rate": 0.001,
|
| 367 |
+
"loss": 1.022,
|
| 368 |
+
"step": 84
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 5.98,
|
| 372 |
+
"grad_norm": 0.2227003127336502,
|
| 373 |
+
"learning_rate": 0.001,
|
| 374 |
+
"loss": 1.0176,
|
| 375 |
+
"step": 86
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 5.98,
|
| 379 |
+
"eval_gen_len": 510.6775147928994,
|
| 380 |
+
"eval_loss": 1.625435709953308,
|
| 381 |
+
"eval_rouge1": 21.1937,
|
| 382 |
+
"eval_rouge2": 6.8813,
|
| 383 |
+
"eval_rougeL": 18.411,
|
| 384 |
+
"eval_rougeLsum": 19.8577,
|
| 385 |
+
"eval_runtime": 1798.5872,
|
| 386 |
+
"eval_samples_per_second": 0.188,
|
| 387 |
+
"eval_steps_per_second": 0.024,
|
| 388 |
+
"step": 86
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 6.12,
|
| 392 |
+
"grad_norm": 0.2959192991256714,
|
| 393 |
+
"learning_rate": 0.001,
|
| 394 |
+
"loss": 0.879,
|
| 395 |
+
"step": 88
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 6.26,
|
| 399 |
+
"grad_norm": 0.33006206154823303,
|
| 400 |
+
"learning_rate": 0.001,
|
| 401 |
+
"loss": 0.8812,
|
| 402 |
+
"step": 90
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 6.4,
|
| 406 |
+
"grad_norm": 0.34284549951553345,
|
| 407 |
+
"learning_rate": 0.001,
|
| 408 |
+
"loss": 0.8742,
|
| 409 |
+
"step": 92
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 6.54,
|
| 413 |
+
"grad_norm": 0.4311819076538086,
|
| 414 |
+
"learning_rate": 0.001,
|
| 415 |
+
"loss": 0.8357,
|
| 416 |
+
"step": 94
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 6.68,
|
| 420 |
+
"grad_norm": 0.5699031352996826,
|
| 421 |
+
"learning_rate": 0.001,
|
| 422 |
+
"loss": 0.8721,
|
| 423 |
+
"step": 96
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 6.82,
|
| 427 |
+
"grad_norm": 0.39324450492858887,
|
| 428 |
+
"learning_rate": 0.001,
|
| 429 |
+
"loss": 0.8739,
|
| 430 |
+
"step": 98
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 6.96,
|
| 434 |
+
"grad_norm": 0.3442493677139282,
|
| 435 |
+
"learning_rate": 0.001,
|
| 436 |
+
"loss": 0.8472,
|
| 437 |
+
"step": 100
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 6.96,
|
| 441 |
+
"eval_gen_len": 479.9704142011834,
|
| 442 |
+
"eval_loss": 1.6212307214736938,
|
| 443 |
+
"eval_rouge1": 26.1873,
|
| 444 |
+
"eval_rouge2": 9.1581,
|
| 445 |
+
"eval_rougeL": 20.393,
|
| 446 |
+
"eval_rougeLsum": 24.1393,
|
| 447 |
+
"eval_runtime": 1802.4729,
|
| 448 |
+
"eval_samples_per_second": 0.188,
|
| 449 |
+
"eval_steps_per_second": 0.024,
|
| 450 |
+
"step": 100
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 7.1,
|
| 454 |
+
"grad_norm": 0.2600483298301697,
|
| 455 |
+
"learning_rate": 0.001,
|
| 456 |
+
"loss": 0.7568,
|
| 457 |
+
"step": 102
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 7.23,
|
| 461 |
+
"grad_norm": 0.28727108240127563,
|
| 462 |
+
"learning_rate": 0.001,
|
| 463 |
+
"loss": 0.6971,
|
| 464 |
+
"step": 104
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 7.37,
|
| 468 |
+
"grad_norm": 0.3065392076969147,
|
| 469 |
+
"learning_rate": 0.001,
|
| 470 |
+
"loss": 0.6918,
|
| 471 |
+
"step": 106
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 7.51,
|
| 475 |
+
"grad_norm": 0.427791029214859,
|
| 476 |
+
"learning_rate": 0.001,
|
| 477 |
+
"loss": 0.6902,
|
| 478 |
+
"step": 108
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 7.65,
|
| 482 |
+
"grad_norm": 0.48664093017578125,
|
| 483 |
+
"learning_rate": 0.001,
|
| 484 |
+
"loss": 0.7415,
|
| 485 |
+
"step": 110
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 7.79,
|
| 489 |
+
"grad_norm": 0.2857199013233185,
|
| 490 |
+
"learning_rate": 0.001,
|
| 491 |
+
"loss": 0.7442,
|
| 492 |
+
"step": 112
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 7.93,
|
| 496 |
+
"grad_norm": 0.24586661159992218,
|
| 497 |
+
"learning_rate": 0.001,
|
| 498 |
+
"loss": 0.7242,
|
| 499 |
+
"step": 114
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 8.0,
|
| 503 |
+
"eval_gen_len": 506.9112426035503,
|
| 504 |
+
"eval_loss": 1.723126769065857,
|
| 505 |
+
"eval_rouge1": 23.5881,
|
| 506 |
+
"eval_rouge2": 7.8961,
|
| 507 |
+
"eval_rougeL": 18.7014,
|
| 508 |
+
"eval_rougeLsum": 22.2999,
|
| 509 |
+
"eval_runtime": 1807.8192,
|
| 510 |
+
"eval_samples_per_second": 0.187,
|
| 511 |
+
"eval_steps_per_second": 0.024,
|
| 512 |
+
"step": 115
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 8.07,
|
| 516 |
+
"grad_norm": 0.21033655107021332,
|
| 517 |
+
"learning_rate": 0.001,
|
| 518 |
+
"loss": 0.6797,
|
| 519 |
+
"step": 116
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 8.21,
|
| 523 |
+
"grad_norm": 0.22591687738895416,
|
| 524 |
+
"learning_rate": 0.001,
|
| 525 |
+
"loss": 0.5446,
|
| 526 |
+
"step": 118
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 8.35,
|
| 530 |
+
"grad_norm": 0.20658165216445923,
|
| 531 |
+
"learning_rate": 0.001,
|
| 532 |
+
"loss": 0.5545,
|
| 533 |
+
"step": 120
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 8.49,
|
| 537 |
+
"grad_norm": 0.29855239391326904,
|
| 538 |
+
"learning_rate": 0.001,
|
| 539 |
+
"loss": 0.6124,
|
| 540 |
+
"step": 122
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 8.63,
|
| 544 |
+
"grad_norm": 0.3976292312145233,
|
| 545 |
+
"learning_rate": 0.001,
|
| 546 |
+
"loss": 0.6052,
|
| 547 |
+
"step": 124
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 8.77,
|
| 551 |
+
"grad_norm": 0.27770739793777466,
|
| 552 |
+
"learning_rate": 0.001,
|
| 553 |
+
"loss": 0.5755,
|
| 554 |
+
"step": 126
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 8.9,
|
| 558 |
+
"grad_norm": 0.2741471529006958,
|
| 559 |
+
"learning_rate": 0.001,
|
| 560 |
+
"loss": 0.5876,
|
| 561 |
+
"step": 128
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 8.97,
|
| 565 |
+
"eval_gen_len": 451.698224852071,
|
| 566 |
+
"eval_loss": 1.9400925636291504,
|
| 567 |
+
"eval_rouge1": 32.1851,
|
| 568 |
+
"eval_rouge2": 12.6426,
|
| 569 |
+
"eval_rougeL": 22.8358,
|
| 570 |
+
"eval_rougeLsum": 30.6718,
|
| 571 |
+
"eval_runtime": 1805.3092,
|
| 572 |
+
"eval_samples_per_second": 0.187,
|
| 573 |
+
"eval_steps_per_second": 0.024,
|
| 574 |
+
"step": 129
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 9.04,
|
| 578 |
+
"grad_norm": 0.25768765807151794,
|
| 579 |
+
"learning_rate": 0.001,
|
| 580 |
+
"loss": 0.5517,
|
| 581 |
+
"step": 130
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 9.18,
|
| 585 |
+
"grad_norm": 0.203142449259758,
|
| 586 |
+
"learning_rate": 0.001,
|
| 587 |
+
"loss": 0.4295,
|
| 588 |
+
"step": 132
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 9.32,
|
| 592 |
+
"grad_norm": 0.29351434111595154,
|
| 593 |
+
"learning_rate": 0.001,
|
| 594 |
+
"loss": 0.493,
|
| 595 |
+
"step": 134
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 9.46,
|
| 599 |
+
"grad_norm": 0.23967808485031128,
|
| 600 |
+
"learning_rate": 0.001,
|
| 601 |
+
"loss": 0.4877,
|
| 602 |
+
"step": 136
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 9.6,
|
| 606 |
+
"grad_norm": 0.21488718688488007,
|
| 607 |
+
"learning_rate": 0.001,
|
| 608 |
+
"loss": 0.4943,
|
| 609 |
+
"step": 138
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 9.74,
|
| 613 |
+
"grad_norm": 0.20587602257728577,
|
| 614 |
+
"learning_rate": 0.001,
|
| 615 |
+
"loss": 0.4729,
|
| 616 |
+
"step": 140
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 9.88,
|
| 620 |
+
"grad_norm": 0.2094978392124176,
|
| 621 |
+
"learning_rate": 0.001,
|
| 622 |
+
"loss": 0.4756,
|
| 623 |
+
"step": 142
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 9.95,
|
| 627 |
+
"eval_gen_len": 455.594674556213,
|
| 628 |
+
"eval_loss": 1.9001177549362183,
|
| 629 |
+
"eval_rouge1": 31.353,
|
| 630 |
+
"eval_rouge2": 12.994,
|
| 631 |
+
"eval_rougeL": 23.1542,
|
| 632 |
+
"eval_rougeLsum": 29.8375,
|
| 633 |
+
"eval_runtime": 1806.0454,
|
| 634 |
+
"eval_samples_per_second": 0.187,
|
| 635 |
+
"eval_steps_per_second": 0.024,
|
| 636 |
+
"step": 143
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 10.02,
|
| 640 |
+
"grad_norm": 0.2443789541721344,
|
| 641 |
+
"learning_rate": 0.001,
|
| 642 |
+
"loss": 0.4707,
|
| 643 |
+
"step": 144
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 10.16,
|
| 647 |
+
"grad_norm": 0.21666786074638367,
|
| 648 |
+
"learning_rate": 0.001,
|
| 649 |
+
"loss": 0.3612,
|
| 650 |
+
"step": 146
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 10.3,
|
| 654 |
+
"grad_norm": 0.20268017053604126,
|
| 655 |
+
"learning_rate": 0.001,
|
| 656 |
+
"loss": 0.3739,
|
| 657 |
+
"step": 148
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 10.43,
|
| 661 |
+
"grad_norm": 0.22428925335407257,
|
| 662 |
+
"learning_rate": 0.001,
|
| 663 |
+
"loss": 0.382,
|
| 664 |
+
"step": 150
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 10.57,
|
| 668 |
+
"grad_norm": 0.21844923496246338,
|
| 669 |
+
"learning_rate": 0.001,
|
| 670 |
+
"loss": 0.3623,
|
| 671 |
+
"step": 152
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 10.71,
|
| 675 |
+
"grad_norm": 0.2675388753414154,
|
| 676 |
+
"learning_rate": 0.001,
|
| 677 |
+
"loss": 0.3674,
|
| 678 |
+
"step": 154
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 10.85,
|
| 682 |
+
"grad_norm": 0.2905120849609375,
|
| 683 |
+
"learning_rate": 0.001,
|
| 684 |
+
"loss": 0.39,
|
| 685 |
+
"step": 156
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 10.99,
|
| 689 |
+
"grad_norm": 0.27420204877853394,
|
| 690 |
+
"learning_rate": 0.001,
|
| 691 |
+
"loss": 0.4042,
|
| 692 |
+
"step": 158
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 10.99,
|
| 696 |
+
"eval_gen_len": 497.53550295857985,
|
| 697 |
+
"eval_loss": 2.1294684410095215,
|
| 698 |
+
"eval_rouge1": 28.6425,
|
| 699 |
+
"eval_rouge2": 11.8399,
|
| 700 |
+
"eval_rougeL": 21.3847,
|
| 701 |
+
"eval_rougeLsum": 27.0508,
|
| 702 |
+
"eval_runtime": 1807.4153,
|
| 703 |
+
"eval_samples_per_second": 0.187,
|
| 704 |
+
"eval_steps_per_second": 0.024,
|
| 705 |
+
"step": 158
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 11.13,
|
| 709 |
+
"grad_norm": 0.26691916584968567,
|
| 710 |
+
"learning_rate": 0.001,
|
| 711 |
+
"loss": 0.3127,
|
| 712 |
+
"step": 160
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 11.27,
|
| 716 |
+
"grad_norm": 0.3042663335800171,
|
| 717 |
+
"learning_rate": 0.001,
|
| 718 |
+
"loss": 0.305,
|
| 719 |
+
"step": 162
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 11.41,
|
| 723 |
+
"grad_norm": 0.26255106925964355,
|
| 724 |
+
"learning_rate": 0.001,
|
| 725 |
+
"loss": 0.3133,
|
| 726 |
+
"step": 164
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 11.55,
|
| 730 |
+
"grad_norm": 0.23816817998886108,
|
| 731 |
+
"learning_rate": 0.001,
|
| 732 |
+
"loss": 0.3118,
|
| 733 |
+
"step": 166
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 11.69,
|
| 737 |
+
"grad_norm": 0.22553777694702148,
|
| 738 |
+
"learning_rate": 0.001,
|
| 739 |
+
"loss": 0.3073,
|
| 740 |
+
"step": 168
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 11.83,
|
| 744 |
+
"grad_norm": 0.2234884351491928,
|
| 745 |
+
"learning_rate": 0.001,
|
| 746 |
+
"loss": 0.3346,
|
| 747 |
+
"step": 170
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 11.97,
|
| 751 |
+
"grad_norm": 0.18143154680728912,
|
| 752 |
+
"learning_rate": 0.001,
|
| 753 |
+
"loss": 0.3292,
|
| 754 |
+
"step": 172
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 11.97,
|
| 758 |
+
"eval_gen_len": 478.81065088757396,
|
| 759 |
+
"eval_loss": 2.244086503982544,
|
| 760 |
+
"eval_rouge1": 31.8393,
|
| 761 |
+
"eval_rouge2": 13.1308,
|
| 762 |
+
"eval_rougeL": 22.135,
|
| 763 |
+
"eval_rougeLsum": 30.5866,
|
| 764 |
+
"eval_runtime": 1798.3958,
|
| 765 |
+
"eval_samples_per_second": 0.188,
|
| 766 |
+
"eval_steps_per_second": 0.024,
|
| 767 |
+
"step": 172
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 12.1,
|
| 771 |
+
"grad_norm": 0.24745677411556244,
|
| 772 |
+
"learning_rate": 0.001,
|
| 773 |
+
"loss": 0.2539,
|
| 774 |
+
"step": 174
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 12.24,
|
| 778 |
+
"grad_norm": 0.26513755321502686,
|
| 779 |
+
"learning_rate": 0.001,
|
| 780 |
+
"loss": 0.2588,
|
| 781 |
+
"step": 176
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 12.38,
|
| 785 |
+
"grad_norm": 0.20156317949295044,
|
| 786 |
+
"learning_rate": 0.001,
|
| 787 |
+
"loss": 0.2537,
|
| 788 |
+
"step": 178
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 12.52,
|
| 792 |
+
"grad_norm": 0.21362556517124176,
|
| 793 |
+
"learning_rate": 0.001,
|
| 794 |
+
"loss": 0.2812,
|
| 795 |
+
"step": 180
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 12.66,
|
| 799 |
+
"grad_norm": 0.5383086800575256,
|
| 800 |
+
"learning_rate": 0.001,
|
| 801 |
+
"loss": 0.2594,
|
| 802 |
+
"step": 182
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 12.8,
|
| 806 |
+
"grad_norm": 0.2891131639480591,
|
| 807 |
+
"learning_rate": 0.001,
|
| 808 |
+
"loss": 0.2629,
|
| 809 |
+
"step": 184
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 12.94,
|
| 813 |
+
"grad_norm": 0.265836238861084,
|
| 814 |
+
"learning_rate": 0.001,
|
| 815 |
+
"loss": 0.2812,
|
| 816 |
+
"step": 186
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 12.94,
|
| 820 |
+
"eval_gen_len": 429.99112426035504,
|
| 821 |
+
"eval_loss": 2.3464245796203613,
|
| 822 |
+
"eval_rouge1": 34.4102,
|
| 823 |
+
"eval_rouge2": 14.3607,
|
| 824 |
+
"eval_rougeL": 23.8634,
|
| 825 |
+
"eval_rougeLsum": 32.9732,
|
| 826 |
+
"eval_runtime": 1798.2194,
|
| 827 |
+
"eval_samples_per_second": 0.188,
|
| 828 |
+
"eval_steps_per_second": 0.024,
|
| 829 |
+
"step": 186
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 13.08,
|
| 833 |
+
"grad_norm": 0.2541401982307434,
|
| 834 |
+
"learning_rate": 0.001,
|
| 835 |
+
"loss": 0.2283,
|
| 836 |
+
"step": 188
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 13.22,
|
| 840 |
+
"grad_norm": 9.848714828491211,
|
| 841 |
+
"learning_rate": 0.001,
|
| 842 |
+
"loss": 0.206,
|
| 843 |
+
"step": 190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 13.36,
|
| 847 |
+
"grad_norm": 0.4088878333568573,
|
| 848 |
+
"learning_rate": 0.001,
|
| 849 |
+
"loss": 0.2014,
|
| 850 |
+
"step": 192
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 13.5,
|
| 854 |
+
"grad_norm": 0.4533099830150604,
|
| 855 |
+
"learning_rate": 0.001,
|
| 856 |
+
"loss": 0.2292,
|
| 857 |
+
"step": 194
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 13.63,
|
| 861 |
+
"grad_norm": 0.28066885471343994,
|
| 862 |
+
"learning_rate": 0.001,
|
| 863 |
+
"loss": 0.2202,
|
| 864 |
+
"step": 196
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 13.77,
|
| 868 |
+
"grad_norm": 0.38810494542121887,
|
| 869 |
+
"learning_rate": 0.001,
|
| 870 |
+
"loss": 0.2278,
|
| 871 |
+
"step": 198
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 13.91,
|
| 875 |
+
"grad_norm": 0.2568497657775879,
|
| 876 |
+
"learning_rate": 0.001,
|
| 877 |
+
"loss": 0.2443,
|
| 878 |
+
"step": 200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 13.98,
|
| 882 |
+
"eval_gen_len": 392.53846153846155,
|
| 883 |
+
"eval_loss": 2.2002713680267334,
|
| 884 |
+
"eval_rouge1": 34.8239,
|
| 885 |
+
"eval_rouge2": 14.8042,
|
| 886 |
+
"eval_rougeL": 25.2438,
|
| 887 |
+
"eval_rougeLsum": 33.0469,
|
| 888 |
+
"eval_runtime": 1797.5392,
|
| 889 |
+
"eval_samples_per_second": 0.188,
|
| 890 |
+
"eval_steps_per_second": 0.024,
|
| 891 |
+
"step": 201
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 14.05,
|
| 895 |
+
"grad_norm": 0.279291570186615,
|
| 896 |
+
"learning_rate": 0.001,
|
| 897 |
+
"loss": 0.2362,
|
| 898 |
+
"step": 202
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 14.19,
|
| 902 |
+
"grad_norm": 0.18151430785655975,
|
| 903 |
+
"learning_rate": 0.001,
|
| 904 |
+
"loss": 0.1807,
|
| 905 |
+
"step": 204
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 14.33,
|
| 909 |
+
"grad_norm": 0.2227843850851059,
|
| 910 |
+
"learning_rate": 0.001,
|
| 911 |
+
"loss": 0.1708,
|
| 912 |
+
"step": 206
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 14.47,
|
| 916 |
+
"grad_norm": 0.2937067151069641,
|
| 917 |
+
"learning_rate": 0.001,
|
| 918 |
+
"loss": 0.1818,
|
| 919 |
+
"step": 208
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 14.61,
|
| 923 |
+
"grad_norm": 0.3238927125930786,
|
| 924 |
+
"learning_rate": 0.001,
|
| 925 |
+
"loss": 0.1958,
|
| 926 |
+
"step": 210
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 14.61,
|
| 930 |
+
"eval_gen_len": 503.5769230769231,
|
| 931 |
+
"eval_loss": 2.5840089321136475,
|
| 932 |
+
"eval_rouge1": 29.7482,
|
| 933 |
+
"eval_rouge2": 12.0072,
|
| 934 |
+
"eval_rougeL": 21.348,
|
| 935 |
+
"eval_rougeLsum": 28.5849,
|
| 936 |
+
"eval_runtime": 1799.2535,
|
| 937 |
+
"eval_samples_per_second": 0.188,
|
| 938 |
+
"eval_steps_per_second": 0.024,
|
| 939 |
+
"step": 210
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 14.61,
|
| 943 |
+
"step": 210,
|
| 944 |
+
"total_flos": 3.6715210940733604e+18,
|
| 945 |
+
"train_loss": 0.9784720075981957,
|
| 946 |
+
"train_runtime": 78497.3694,
|
| 947 |
+
"train_samples_per_second": 0.702,
|
| 948 |
+
"train_steps_per_second": 0.003
|
| 949 |
+
}
|
| 950 |
+
],
|
| 951 |
+
"logging_steps": 2,
|
| 952 |
+
"max_steps": 210,
|
| 953 |
+
"num_input_tokens_seen": 0,
|
| 954 |
+
"num_train_epochs": 15,
|
| 955 |
+
"save_steps": 500,
|
| 956 |
+
"total_flos": 3.6715210940733604e+18,
|
| 957 |
+
"train_batch_size": 8,
|
| 958 |
+
"trial_name": null,
|
| 959 |
+
"trial_params": null
|
| 960 |
+
}
|