End of training
Browse files- README.md +14 -2
- all_results.json +21 -0
- eval_results.json +16 -0
- train_results.json +8 -0
- trainer_state.json +806 -0
README.md
CHANGED
|
@@ -3,11 +3,23 @@ license: mit
|
|
| 3 |
base_model: gpt2-xl
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
| 8 |
model-index:
|
| 9 |
- name: lmind_hotpot_train300_eval100_v1_recite_qa_gpt2-xl
|
| 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 |
# lmind_hotpot_train300_eval100_v1_recite_qa_gpt2-xl
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [gpt2-xl](https://huggingface.co/gpt2-xl) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.4199
|
| 21 |
- Accuracy: 0.6908
|
|
|
|
| 3 |
base_model: gpt2-xl
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- tyzhu/lmind_hotpot_train300_eval100_v1_recite_qa
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
| 10 |
model-index:
|
| 11 |
- name: lmind_hotpot_train300_eval100_v1_recite_qa_gpt2-xl
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Causal Language Modeling
|
| 15 |
+
type: text-generation
|
| 16 |
+
dataset:
|
| 17 |
+
name: tyzhu/lmind_hotpot_train300_eval100_v1_recite_qa
|
| 18 |
+
type: tyzhu/lmind_hotpot_train300_eval100_v1_recite_qa
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Accuracy
|
| 21 |
+
type: accuracy
|
| 22 |
+
value: 0.6908442503639011
|
| 23 |
---
|
| 24 |
|
| 25 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 27 |
|
| 28 |
# lmind_hotpot_train300_eval100_v1_recite_qa_gpt2-xl
|
| 29 |
|
| 30 |
+
This model is a fine-tuned version of [gpt2-xl](https://huggingface.co/gpt2-xl) on the tyzhu/lmind_hotpot_train300_eval100_v1_recite_qa dataset.
|
| 31 |
It achieves the following results on the evaluation set:
|
| 32 |
- Loss: 0.4199
|
| 33 |
- Accuracy: 0.6908
|
all_results.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.6908442503639011,
|
| 4 |
+
"eval_exact_match": 19.0,
|
| 5 |
+
"eval_f1": 24.855555555555558,
|
| 6 |
+
"eval_loss": 0.4198664426803589,
|
| 7 |
+
"eval_qa_bleu": 14.403045967487259,
|
| 8 |
+
"eval_qa_exact_match": 0.18,
|
| 9 |
+
"eval_recite_bleu": 45.62406704093971,
|
| 10 |
+
"eval_recite_exact_match": 0.02,
|
| 11 |
+
"eval_runtime": 3.2494,
|
| 12 |
+
"eval_samples": 100,
|
| 13 |
+
"eval_samples_per_second": 30.775,
|
| 14 |
+
"eval_steps_per_second": 2.154,
|
| 15 |
+
"perplexity": 1.5217583000861028,
|
| 16 |
+
"train_loss": 0.7305109727641811,
|
| 17 |
+
"train_runtime": 1813.2807,
|
| 18 |
+
"train_samples": 1097,
|
| 19 |
+
"train_samples_per_second": 6.05,
|
| 20 |
+
"train_steps_per_second": 0.381
|
| 21 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.6908442503639011,
|
| 4 |
+
"eval_exact_match": 19.0,
|
| 5 |
+
"eval_f1": 24.855555555555558,
|
| 6 |
+
"eval_loss": 0.4198664426803589,
|
| 7 |
+
"eval_qa_bleu": 14.403045967487259,
|
| 8 |
+
"eval_qa_exact_match": 0.18,
|
| 9 |
+
"eval_recite_bleu": 45.62406704093971,
|
| 10 |
+
"eval_recite_exact_match": 0.02,
|
| 11 |
+
"eval_runtime": 3.2494,
|
| 12 |
+
"eval_samples": 100,
|
| 13 |
+
"eval_samples_per_second": 30.775,
|
| 14 |
+
"eval_steps_per_second": 2.154,
|
| 15 |
+
"perplexity": 1.5217583000861028
|
| 16 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 0.7305109727641811,
|
| 4 |
+
"train_runtime": 1813.2807,
|
| 5 |
+
"train_samples": 1097,
|
| 6 |
+
"train_samples_per_second": 6.05,
|
| 7 |
+
"train_steps_per_second": 0.381
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,806 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 690,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1,
|
| 13 |
+
"learning_rate": 6e-06,
|
| 14 |
+
"loss": 7.4689,
|
| 15 |
+
"step": 7
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.2,
|
| 19 |
+
"learning_rate": 1.2e-05,
|
| 20 |
+
"loss": 5.1854,
|
| 21 |
+
"step": 14
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.3,
|
| 25 |
+
"learning_rate": 1.8e-05,
|
| 26 |
+
"loss": 3.962,
|
| 27 |
+
"step": 21
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.41,
|
| 31 |
+
"learning_rate": 2.4e-05,
|
| 32 |
+
"loss": 2.9971,
|
| 33 |
+
"step": 28
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.51,
|
| 37 |
+
"learning_rate": 3e-05,
|
| 38 |
+
"loss": 2.342,
|
| 39 |
+
"step": 35
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.61,
|
| 43 |
+
"learning_rate": 2.9679389312977098e-05,
|
| 44 |
+
"loss": 2.2103,
|
| 45 |
+
"step": 42
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.71,
|
| 49 |
+
"learning_rate": 2.93587786259542e-05,
|
| 50 |
+
"loss": 2.1337,
|
| 51 |
+
"step": 49
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.81,
|
| 55 |
+
"learning_rate": 2.9038167938931298e-05,
|
| 56 |
+
"loss": 2.0715,
|
| 57 |
+
"step": 56
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.91,
|
| 61 |
+
"learning_rate": 2.8717557251908395e-05,
|
| 62 |
+
"loss": 1.9548,
|
| 63 |
+
"step": 63
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.0,
|
| 67 |
+
"eval_accuracy": 0.5812809315866084,
|
| 68 |
+
"eval_loss": 1.7231289148330688,
|
| 69 |
+
"eval_runtime": 9.0359,
|
| 70 |
+
"eval_samples_per_second": 11.067,
|
| 71 |
+
"eval_steps_per_second": 0.775,
|
| 72 |
+
"step": 69
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.0,
|
| 76 |
+
"eval_exact_match": 6.0,
|
| 77 |
+
"eval_f1": 10.928571428571429,
|
| 78 |
+
"eval_qa_bleu": 3.6846142429687943,
|
| 79 |
+
"eval_qa_exact_match": 0.06,
|
| 80 |
+
"eval_recite_bleu": 12.518352954015292,
|
| 81 |
+
"eval_recite_exact_match": 0.0,
|
| 82 |
+
"step": 69
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 1.01,
|
| 86 |
+
"learning_rate": 2.8396946564885498e-05,
|
| 87 |
+
"loss": 1.9164,
|
| 88 |
+
"step": 70
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.12,
|
| 92 |
+
"learning_rate": 2.8076335877862595e-05,
|
| 93 |
+
"loss": 1.734,
|
| 94 |
+
"step": 77
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.22,
|
| 98 |
+
"learning_rate": 2.7755725190839695e-05,
|
| 99 |
+
"loss": 1.5936,
|
| 100 |
+
"step": 84
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 1.32,
|
| 104 |
+
"learning_rate": 2.7435114503816795e-05,
|
| 105 |
+
"loss": 1.4865,
|
| 106 |
+
"step": 91
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.42,
|
| 110 |
+
"learning_rate": 2.7114503816793892e-05,
|
| 111 |
+
"loss": 1.5131,
|
| 112 |
+
"step": 98
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.52,
|
| 116 |
+
"learning_rate": 2.6793893129770996e-05,
|
| 117 |
+
"loss": 1.4826,
|
| 118 |
+
"step": 105
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 1.62,
|
| 122 |
+
"learning_rate": 2.6473282442748093e-05,
|
| 123 |
+
"loss": 1.5076,
|
| 124 |
+
"step": 112
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 1.72,
|
| 128 |
+
"learning_rate": 2.615267175572519e-05,
|
| 129 |
+
"loss": 1.3919,
|
| 130 |
+
"step": 119
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.83,
|
| 134 |
+
"learning_rate": 2.5832061068702293e-05,
|
| 135 |
+
"loss": 1.458,
|
| 136 |
+
"step": 126
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.93,
|
| 140 |
+
"learning_rate": 2.551145038167939e-05,
|
| 141 |
+
"loss": 1.3306,
|
| 142 |
+
"step": 133
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 2.0,
|
| 146 |
+
"eval_accuracy": 0.6136244541484717,
|
| 147 |
+
"eval_loss": 1.2326337099075317,
|
| 148 |
+
"eval_runtime": 3.2768,
|
| 149 |
+
"eval_samples_per_second": 30.517,
|
| 150 |
+
"eval_steps_per_second": 2.136,
|
| 151 |
+
"step": 138
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.0,
|
| 155 |
+
"eval_exact_match": 9.0,
|
| 156 |
+
"eval_f1": 18.469047619047622,
|
| 157 |
+
"eval_qa_bleu": 10.388842490007553,
|
| 158 |
+
"eval_qa_exact_match": 0.08,
|
| 159 |
+
"eval_recite_bleu": 18.414656875876464,
|
| 160 |
+
"eval_recite_exact_match": 0.0,
|
| 161 |
+
"step": 138
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 2.03,
|
| 165 |
+
"learning_rate": 2.5190839694656487e-05,
|
| 166 |
+
"loss": 1.2798,
|
| 167 |
+
"step": 140
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.13,
|
| 171 |
+
"learning_rate": 2.487022900763359e-05,
|
| 172 |
+
"loss": 0.911,
|
| 173 |
+
"step": 147
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 2.23,
|
| 177 |
+
"learning_rate": 2.4549618320610687e-05,
|
| 178 |
+
"loss": 0.9711,
|
| 179 |
+
"step": 154
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.33,
|
| 183 |
+
"learning_rate": 2.4229007633587784e-05,
|
| 184 |
+
"loss": 0.9432,
|
| 185 |
+
"step": 161
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 2.43,
|
| 189 |
+
"learning_rate": 2.3908396946564887e-05,
|
| 190 |
+
"loss": 0.8923,
|
| 191 |
+
"step": 168
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 2.54,
|
| 195 |
+
"learning_rate": 2.3587786259541984e-05,
|
| 196 |
+
"loss": 0.8927,
|
| 197 |
+
"step": 175
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 2.64,
|
| 201 |
+
"learning_rate": 2.3267175572519084e-05,
|
| 202 |
+
"loss": 0.8711,
|
| 203 |
+
"step": 182
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 2.74,
|
| 207 |
+
"learning_rate": 2.2946564885496185e-05,
|
| 208 |
+
"loss": 0.8087,
|
| 209 |
+
"step": 189
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.84,
|
| 213 |
+
"learning_rate": 2.262595419847328e-05,
|
| 214 |
+
"loss": 0.8442,
|
| 215 |
+
"step": 196
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 2.94,
|
| 219 |
+
"learning_rate": 2.2305343511450385e-05,
|
| 220 |
+
"loss": 0.8853,
|
| 221 |
+
"step": 203
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 3.0,
|
| 225 |
+
"eval_accuracy": 0.6420524017467248,
|
| 226 |
+
"eval_loss": 0.8816230297088623,
|
| 227 |
+
"eval_runtime": 3.3365,
|
| 228 |
+
"eval_samples_per_second": 29.971,
|
| 229 |
+
"eval_steps_per_second": 2.098,
|
| 230 |
+
"step": 207
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 3.0,
|
| 234 |
+
"eval_exact_match": 9.0,
|
| 235 |
+
"eval_f1": 16.49920634920635,
|
| 236 |
+
"eval_qa_bleu": 8.802610551775999,
|
| 237 |
+
"eval_qa_exact_match": 0.09,
|
| 238 |
+
"eval_recite_bleu": 22.99064826237382,
|
| 239 |
+
"eval_recite_exact_match": 0.0,
|
| 240 |
+
"step": 207
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.04,
|
| 244 |
+
"learning_rate": 2.198473282442748e-05,
|
| 245 |
+
"loss": 0.6919,
|
| 246 |
+
"step": 210
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.14,
|
| 250 |
+
"learning_rate": 2.166412213740458e-05,
|
| 251 |
+
"loss": 0.5836,
|
| 252 |
+
"step": 217
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 3.25,
|
| 256 |
+
"learning_rate": 2.1343511450381682e-05,
|
| 257 |
+
"loss": 0.5566,
|
| 258 |
+
"step": 224
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 3.35,
|
| 262 |
+
"learning_rate": 2.102290076335878e-05,
|
| 263 |
+
"loss": 0.5625,
|
| 264 |
+
"step": 231
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.45,
|
| 268 |
+
"learning_rate": 2.0702290076335876e-05,
|
| 269 |
+
"loss": 0.5301,
|
| 270 |
+
"step": 238
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 3.55,
|
| 274 |
+
"learning_rate": 2.038167938931298e-05,
|
| 275 |
+
"loss": 0.5601,
|
| 276 |
+
"step": 245
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 3.65,
|
| 280 |
+
"learning_rate": 2.0061068702290076e-05,
|
| 281 |
+
"loss": 0.5088,
|
| 282 |
+
"step": 252
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 3.75,
|
| 286 |
+
"learning_rate": 1.9740458015267176e-05,
|
| 287 |
+
"loss": 0.5661,
|
| 288 |
+
"step": 259
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.86,
|
| 292 |
+
"learning_rate": 1.9419847328244276e-05,
|
| 293 |
+
"loss": 0.5007,
|
| 294 |
+
"step": 266
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 3.96,
|
| 298 |
+
"learning_rate": 1.9099236641221373e-05,
|
| 299 |
+
"loss": 0.5181,
|
| 300 |
+
"step": 273
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 4.0,
|
| 304 |
+
"eval_accuracy": 0.6637554585152838,
|
| 305 |
+
"eval_loss": 0.6443885564804077,
|
| 306 |
+
"eval_runtime": 3.1883,
|
| 307 |
+
"eval_samples_per_second": 31.365,
|
| 308 |
+
"eval_steps_per_second": 2.196,
|
| 309 |
+
"step": 276
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 4.0,
|
| 313 |
+
"eval_exact_match": 9.0,
|
| 314 |
+
"eval_f1": 15.252380952380953,
|
| 315 |
+
"eval_qa_bleu": 4.187460029928345,
|
| 316 |
+
"eval_qa_exact_match": 0.09,
|
| 317 |
+
"eval_recite_bleu": 24.09222861395294,
|
| 318 |
+
"eval_recite_exact_match": 0.0,
|
| 319 |
+
"step": 276
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 4.06,
|
| 323 |
+
"learning_rate": 1.8778625954198473e-05,
|
| 324 |
+
"loss": 0.4274,
|
| 325 |
+
"step": 280
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 4.16,
|
| 329 |
+
"learning_rate": 1.8458015267175574e-05,
|
| 330 |
+
"loss": 0.3626,
|
| 331 |
+
"step": 287
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 4.26,
|
| 335 |
+
"learning_rate": 1.813740458015267e-05,
|
| 336 |
+
"loss": 0.3519,
|
| 337 |
+
"step": 294
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 4.36,
|
| 341 |
+
"learning_rate": 1.7816793893129774e-05,
|
| 342 |
+
"loss": 0.3387,
|
| 343 |
+
"step": 301
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 4.46,
|
| 347 |
+
"learning_rate": 1.749618320610687e-05,
|
| 348 |
+
"loss": 0.331,
|
| 349 |
+
"step": 308
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 4.57,
|
| 353 |
+
"learning_rate": 1.7175572519083968e-05,
|
| 354 |
+
"loss": 0.3131,
|
| 355 |
+
"step": 315
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 4.67,
|
| 359 |
+
"learning_rate": 1.685496183206107e-05,
|
| 360 |
+
"loss": 0.3166,
|
| 361 |
+
"step": 322
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 4.77,
|
| 365 |
+
"learning_rate": 1.6534351145038168e-05,
|
| 366 |
+
"loss": 0.3555,
|
| 367 |
+
"step": 329
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 4.87,
|
| 371 |
+
"learning_rate": 1.6213740458015265e-05,
|
| 372 |
+
"loss": 0.3394,
|
| 373 |
+
"step": 336
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 4.97,
|
| 377 |
+
"learning_rate": 1.5893129770992368e-05,
|
| 378 |
+
"loss": 0.3236,
|
| 379 |
+
"step": 343
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 5.0,
|
| 383 |
+
"eval_accuracy": 0.6771324599708879,
|
| 384 |
+
"eval_loss": 0.5304832458496094,
|
| 385 |
+
"eval_runtime": 3.2706,
|
| 386 |
+
"eval_samples_per_second": 30.576,
|
| 387 |
+
"eval_steps_per_second": 2.14,
|
| 388 |
+
"step": 345
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 5.0,
|
| 392 |
+
"eval_exact_match": 6.0,
|
| 393 |
+
"eval_f1": 12.238095238095239,
|
| 394 |
+
"eval_qa_bleu": 3.673522350616593,
|
| 395 |
+
"eval_qa_exact_match": 0.06,
|
| 396 |
+
"eval_recite_bleu": 28.72242820390079,
|
| 397 |
+
"eval_recite_exact_match": 0.0,
|
| 398 |
+
"step": 345
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 5.07,
|
| 402 |
+
"learning_rate": 1.5572519083969465e-05,
|
| 403 |
+
"loss": 0.2619,
|
| 404 |
+
"step": 350
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 5.17,
|
| 408 |
+
"learning_rate": 1.5251908396946567e-05,
|
| 409 |
+
"loss": 0.2471,
|
| 410 |
+
"step": 357
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 5.28,
|
| 414 |
+
"learning_rate": 1.4931297709923665e-05,
|
| 415 |
+
"loss": 0.2407,
|
| 416 |
+
"step": 364
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 5.38,
|
| 420 |
+
"learning_rate": 1.4610687022900764e-05,
|
| 421 |
+
"loss": 0.2503,
|
| 422 |
+
"step": 371
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 5.48,
|
| 426 |
+
"learning_rate": 1.4290076335877862e-05,
|
| 427 |
+
"loss": 0.225,
|
| 428 |
+
"step": 378
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 5.58,
|
| 432 |
+
"learning_rate": 1.3969465648854963e-05,
|
| 433 |
+
"loss": 0.2486,
|
| 434 |
+
"step": 385
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 5.68,
|
| 438 |
+
"learning_rate": 1.3648854961832061e-05,
|
| 439 |
+
"loss": 0.2366,
|
| 440 |
+
"step": 392
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 5.78,
|
| 444 |
+
"learning_rate": 1.332824427480916e-05,
|
| 445 |
+
"loss": 0.2371,
|
| 446 |
+
"step": 399
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 5.88,
|
| 450 |
+
"learning_rate": 1.300763358778626e-05,
|
| 451 |
+
"loss": 0.2308,
|
| 452 |
+
"step": 406
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 5.99,
|
| 456 |
+
"learning_rate": 1.268702290076336e-05,
|
| 457 |
+
"loss": 0.2371,
|
| 458 |
+
"step": 413
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 6.0,
|
| 462 |
+
"eval_accuracy": 0.6848471615720524,
|
| 463 |
+
"eval_loss": 0.4593181312084198,
|
| 464 |
+
"eval_runtime": 3.3218,
|
| 465 |
+
"eval_samples_per_second": 30.104,
|
| 466 |
+
"eval_steps_per_second": 2.107,
|
| 467 |
+
"step": 414
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 6.0,
|
| 471 |
+
"eval_exact_match": 12.0,
|
| 472 |
+
"eval_f1": 19.94920634920635,
|
| 473 |
+
"eval_qa_bleu": 7.958666643852284,
|
| 474 |
+
"eval_qa_exact_match": 0.12,
|
| 475 |
+
"eval_recite_bleu": 36.347059921630894,
|
| 476 |
+
"eval_recite_exact_match": 0.0,
|
| 477 |
+
"step": 414
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 6.09,
|
| 481 |
+
"learning_rate": 1.2366412213740458e-05,
|
| 482 |
+
"loss": 0.1793,
|
| 483 |
+
"step": 420
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 6.19,
|
| 487 |
+
"learning_rate": 1.2045801526717557e-05,
|
| 488 |
+
"loss": 0.179,
|
| 489 |
+
"step": 427
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 6.29,
|
| 493 |
+
"learning_rate": 1.1725190839694657e-05,
|
| 494 |
+
"loss": 0.1729,
|
| 495 |
+
"step": 434
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 6.39,
|
| 499 |
+
"learning_rate": 1.1404580152671756e-05,
|
| 500 |
+
"loss": 0.1748,
|
| 501 |
+
"step": 441
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 6.49,
|
| 505 |
+
"learning_rate": 1.1083969465648856e-05,
|
| 506 |
+
"loss": 0.1847,
|
| 507 |
+
"step": 448
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 6.59,
|
| 511 |
+
"learning_rate": 1.0763358778625954e-05,
|
| 512 |
+
"loss": 0.1823,
|
| 513 |
+
"step": 455
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 6.7,
|
| 517 |
+
"learning_rate": 1.0442748091603054e-05,
|
| 518 |
+
"loss": 0.1825,
|
| 519 |
+
"step": 462
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 6.8,
|
| 523 |
+
"learning_rate": 1.0122137404580153e-05,
|
| 524 |
+
"loss": 0.1651,
|
| 525 |
+
"step": 469
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 6.9,
|
| 529 |
+
"learning_rate": 9.801526717557251e-06,
|
| 530 |
+
"loss": 0.1675,
|
| 531 |
+
"step": 476
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 7.0,
|
| 535 |
+
"learning_rate": 9.480916030534352e-06,
|
| 536 |
+
"loss": 0.1839,
|
| 537 |
+
"step": 483
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 7.0,
|
| 541 |
+
"eval_accuracy": 0.6880931586608442,
|
| 542 |
+
"eval_loss": 0.43847352266311646,
|
| 543 |
+
"eval_runtime": 4.5343,
|
| 544 |
+
"eval_samples_per_second": 22.054,
|
| 545 |
+
"eval_steps_per_second": 1.544,
|
| 546 |
+
"step": 483
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 7.0,
|
| 550 |
+
"eval_exact_match": 10.0,
|
| 551 |
+
"eval_f1": 16.56111111111111,
|
| 552 |
+
"eval_qa_bleu": 5.445893929820159,
|
| 553 |
+
"eval_qa_exact_match": 0.1,
|
| 554 |
+
"eval_recite_bleu": 38.055647893528544,
|
| 555 |
+
"eval_recite_exact_match": 0.0,
|
| 556 |
+
"step": 483
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 7.1,
|
| 560 |
+
"learning_rate": 9.16030534351145e-06,
|
| 561 |
+
"loss": 0.1371,
|
| 562 |
+
"step": 490
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 7.2,
|
| 566 |
+
"learning_rate": 8.83969465648855e-06,
|
| 567 |
+
"loss": 0.1461,
|
| 568 |
+
"step": 497
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 7.3,
|
| 572 |
+
"learning_rate": 8.519083969465649e-06,
|
| 573 |
+
"loss": 0.1343,
|
| 574 |
+
"step": 504
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 7.41,
|
| 578 |
+
"learning_rate": 8.198473282442749e-06,
|
| 579 |
+
"loss": 0.1452,
|
| 580 |
+
"step": 511
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 7.51,
|
| 584 |
+
"learning_rate": 7.877862595419847e-06,
|
| 585 |
+
"loss": 0.1415,
|
| 586 |
+
"step": 518
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 7.61,
|
| 590 |
+
"learning_rate": 7.557251908396946e-06,
|
| 591 |
+
"loss": 0.162,
|
| 592 |
+
"step": 525
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 7.71,
|
| 596 |
+
"learning_rate": 7.236641221374045e-06,
|
| 597 |
+
"loss": 0.1468,
|
| 598 |
+
"step": 532
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 7.81,
|
| 602 |
+
"learning_rate": 6.9160305343511455e-06,
|
| 603 |
+
"loss": 0.14,
|
| 604 |
+
"step": 539
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 7.91,
|
| 608 |
+
"learning_rate": 6.595419847328244e-06,
|
| 609 |
+
"loss": 0.1287,
|
| 610 |
+
"step": 546
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 8.0,
|
| 614 |
+
"eval_accuracy": 0.6898835516739447,
|
| 615 |
+
"eval_loss": 0.42429620027542114,
|
| 616 |
+
"eval_runtime": 3.4153,
|
| 617 |
+
"eval_samples_per_second": 29.28,
|
| 618 |
+
"eval_steps_per_second": 2.05,
|
| 619 |
+
"step": 552
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 8.0,
|
| 623 |
+
"eval_exact_match": 17.0,
|
| 624 |
+
"eval_f1": 21.971428571428575,
|
| 625 |
+
"eval_qa_bleu": 22.723779810738534,
|
| 626 |
+
"eval_qa_exact_match": 0.16,
|
| 627 |
+
"eval_recite_bleu": 42.69152358155121,
|
| 628 |
+
"eval_recite_exact_match": 0.01,
|
| 629 |
+
"step": 552
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 8.01,
|
| 633 |
+
"learning_rate": 6.274809160305344e-06,
|
| 634 |
+
"loss": 0.1479,
|
| 635 |
+
"step": 553
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 8.12,
|
| 639 |
+
"learning_rate": 5.954198473282443e-06,
|
| 640 |
+
"loss": 0.1235,
|
| 641 |
+
"step": 560
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 8.22,
|
| 645 |
+
"learning_rate": 5.633587786259543e-06,
|
| 646 |
+
"loss": 0.1341,
|
| 647 |
+
"step": 567
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 8.32,
|
| 651 |
+
"learning_rate": 5.312977099236641e-06,
|
| 652 |
+
"loss": 0.1178,
|
| 653 |
+
"step": 574
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 8.42,
|
| 657 |
+
"learning_rate": 4.99236641221374e-06,
|
| 658 |
+
"loss": 0.1172,
|
| 659 |
+
"step": 581
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 8.52,
|
| 663 |
+
"learning_rate": 4.67175572519084e-06,
|
| 664 |
+
"loss": 0.1313,
|
| 665 |
+
"step": 588
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 8.62,
|
| 669 |
+
"learning_rate": 4.3511450381679385e-06,
|
| 670 |
+
"loss": 0.1173,
|
| 671 |
+
"step": 595
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 8.72,
|
| 675 |
+
"learning_rate": 4.030534351145039e-06,
|
| 676 |
+
"loss": 0.1148,
|
| 677 |
+
"step": 602
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 8.83,
|
| 681 |
+
"learning_rate": 3.709923664122137e-06,
|
| 682 |
+
"loss": 0.1208,
|
| 683 |
+
"step": 609
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 8.93,
|
| 687 |
+
"learning_rate": 3.3893129770992365e-06,
|
| 688 |
+
"loss": 0.1241,
|
| 689 |
+
"step": 616
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 9.0,
|
| 693 |
+
"eval_accuracy": 0.6904949053857351,
|
| 694 |
+
"eval_loss": 0.4206622242927551,
|
| 695 |
+
"eval_runtime": 3.3293,
|
| 696 |
+
"eval_samples_per_second": 30.036,
|
| 697 |
+
"eval_steps_per_second": 2.103,
|
| 698 |
+
"step": 621
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 9.0,
|
| 702 |
+
"eval_exact_match": 14.0,
|
| 703 |
+
"eval_f1": 20.504761904761907,
|
| 704 |
+
"eval_qa_bleu": 13.525989143509026,
|
| 705 |
+
"eval_qa_exact_match": 0.14,
|
| 706 |
+
"eval_recite_bleu": 45.9618562736209,
|
| 707 |
+
"eval_recite_exact_match": 0.02,
|
| 708 |
+
"step": 621
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 9.03,
|
| 712 |
+
"learning_rate": 3.068702290076336e-06,
|
| 713 |
+
"loss": 0.122,
|
| 714 |
+
"step": 623
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 9.13,
|
| 718 |
+
"learning_rate": 2.748091603053435e-06,
|
| 719 |
+
"loss": 0.1121,
|
| 720 |
+
"step": 630
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 9.23,
|
| 724 |
+
"learning_rate": 2.4274809160305345e-06,
|
| 725 |
+
"loss": 0.1127,
|
| 726 |
+
"step": 637
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 9.33,
|
| 730 |
+
"learning_rate": 2.106870229007634e-06,
|
| 731 |
+
"loss": 0.1148,
|
| 732 |
+
"step": 644
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 9.43,
|
| 736 |
+
"learning_rate": 1.7862595419847328e-06,
|
| 737 |
+
"loss": 0.1123,
|
| 738 |
+
"step": 651
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 9.54,
|
| 742 |
+
"learning_rate": 1.4656488549618321e-06,
|
| 743 |
+
"loss": 0.1094,
|
| 744 |
+
"step": 658
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 9.64,
|
| 748 |
+
"learning_rate": 1.1450381679389313e-06,
|
| 749 |
+
"loss": 0.117,
|
| 750 |
+
"step": 665
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 9.74,
|
| 754 |
+
"learning_rate": 8.244274809160305e-07,
|
| 755 |
+
"loss": 0.118,
|
| 756 |
+
"step": 672
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 9.84,
|
| 760 |
+
"learning_rate": 5.038167938931298e-07,
|
| 761 |
+
"loss": 0.1097,
|
| 762 |
+
"step": 679
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 9.94,
|
| 766 |
+
"learning_rate": 1.8320610687022902e-07,
|
| 767 |
+
"loss": 0.1198,
|
| 768 |
+
"step": 686
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 10.0,
|
| 772 |
+
"eval_accuracy": 0.6908442503639011,
|
| 773 |
+
"eval_loss": 0.4198664426803589,
|
| 774 |
+
"eval_runtime": 3.7502,
|
| 775 |
+
"eval_samples_per_second": 26.666,
|
| 776 |
+
"eval_steps_per_second": 1.867,
|
| 777 |
+
"step": 690
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 10.0,
|
| 781 |
+
"eval_exact_match": 19.0,
|
| 782 |
+
"eval_f1": 24.855555555555558,
|
| 783 |
+
"eval_qa_bleu": 14.403045967487259,
|
| 784 |
+
"eval_qa_exact_match": 0.18,
|
| 785 |
+
"eval_recite_bleu": 45.62406704093971,
|
| 786 |
+
"eval_recite_exact_match": 0.02,
|
| 787 |
+
"step": 690
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 10.0,
|
| 791 |
+
"step": 690,
|
| 792 |
+
"total_flos": 3.18562653216768e+16,
|
| 793 |
+
"train_loss": 0.7305109727641811,
|
| 794 |
+
"train_runtime": 1813.2807,
|
| 795 |
+
"train_samples_per_second": 6.05,
|
| 796 |
+
"train_steps_per_second": 0.381
|
| 797 |
+
}
|
| 798 |
+
],
|
| 799 |
+
"logging_steps": 7,
|
| 800 |
+
"max_steps": 690,
|
| 801 |
+
"num_train_epochs": 10,
|
| 802 |
+
"save_steps": 500,
|
| 803 |
+
"total_flos": 3.18562653216768e+16,
|
| 804 |
+
"trial_name": null,
|
| 805 |
+
"trial_params": null
|
| 806 |
+
}
|