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 +746 -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_nq_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_nq_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: 0.3388
|
| 21 |
- Accuracy: 0.8212
|
|
|
|
| 3 |
base_model: gpt2-xl
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- tyzhu/lmind_nq_train300_eval100_v1_recite_qa
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
| 10 |
model-index:
|
| 11 |
- name: lmind_nq_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_nq_train300_eval100_v1_recite_qa
|
| 18 |
+
type: tyzhu/lmind_nq_train300_eval100_v1_recite_qa
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Accuracy
|
| 21 |
+
type: accuracy
|
| 22 |
+
value: 0.8212156862745098
|
| 23 |
---
|
| 24 |
|
| 25 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 27 |
|
| 28 |
# lmind_nq_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_nq_train300_eval100_v1_recite_qa dataset.
|
| 31 |
It achieves the following results on the evaluation set:
|
| 32 |
- Loss: 0.3388
|
| 33 |
- Accuracy: 0.8212
|
all_results.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8212156862745098,
|
| 4 |
+
"eval_exact_match": 15.0,
|
| 5 |
+
"eval_f1": 20.53333333333333,
|
| 6 |
+
"eval_loss": 0.33876052498817444,
|
| 7 |
+
"eval_qa_bleu": 6.622399363390335,
|
| 8 |
+
"eval_qa_exact_match": 0.13,
|
| 9 |
+
"eval_recite_bleu": 38.34543604737045,
|
| 10 |
+
"eval_recite_exact_match": 0.16,
|
| 11 |
+
"eval_runtime": 2.8203,
|
| 12 |
+
"eval_samples": 100,
|
| 13 |
+
"eval_samples_per_second": 35.457,
|
| 14 |
+
"eval_steps_per_second": 2.482,
|
| 15 |
+
"perplexity": 1.4032072718957205,
|
| 16 |
+
"train_loss": 0.09364926754073663,
|
| 17 |
+
"train_runtime": 1254.2575,
|
| 18 |
+
"train_samples": 692,
|
| 19 |
+
"train_samples_per_second": 5.517,
|
| 20 |
+
"train_steps_per_second": 0.351
|
| 21 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8212156862745098,
|
| 4 |
+
"eval_exact_match": 15.0,
|
| 5 |
+
"eval_f1": 20.53333333333333,
|
| 6 |
+
"eval_loss": 0.33876052498817444,
|
| 7 |
+
"eval_qa_bleu": 6.622399363390335,
|
| 8 |
+
"eval_qa_exact_match": 0.13,
|
| 9 |
+
"eval_recite_bleu": 38.34543604737045,
|
| 10 |
+
"eval_recite_exact_match": 0.16,
|
| 11 |
+
"eval_runtime": 2.8203,
|
| 12 |
+
"eval_samples": 100,
|
| 13 |
+
"eval_samples_per_second": 35.457,
|
| 14 |
+
"eval_steps_per_second": 2.482,
|
| 15 |
+
"perplexity": 1.4032072718957205
|
| 16 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 0.09364926754073663,
|
| 4 |
+
"train_runtime": 1254.2575,
|
| 5 |
+
"train_samples": 692,
|
| 6 |
+
"train_samples_per_second": 5.517,
|
| 7 |
+
"train_steps_per_second": 0.351
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,746 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 440,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11,
|
| 13 |
+
"learning_rate": 3e-05,
|
| 14 |
+
"loss": 0.2793,
|
| 15 |
+
"step": 5
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.23,
|
| 19 |
+
"learning_rate": 3e-05,
|
| 20 |
+
"loss": 0.2329,
|
| 21 |
+
"step": 10
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.34,
|
| 25 |
+
"learning_rate": 3e-05,
|
| 26 |
+
"loss": 0.2271,
|
| 27 |
+
"step": 15
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.45,
|
| 31 |
+
"learning_rate": 3e-05,
|
| 32 |
+
"loss": 0.2263,
|
| 33 |
+
"step": 20
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.57,
|
| 37 |
+
"learning_rate": 3e-05,
|
| 38 |
+
"loss": 0.2071,
|
| 39 |
+
"step": 25
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.68,
|
| 43 |
+
"learning_rate": 3e-05,
|
| 44 |
+
"loss": 0.1903,
|
| 45 |
+
"step": 30
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.8,
|
| 49 |
+
"learning_rate": 3e-05,
|
| 50 |
+
"loss": 0.1967,
|
| 51 |
+
"step": 35
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.91,
|
| 55 |
+
"learning_rate": 3e-05,
|
| 56 |
+
"loss": 0.1943,
|
| 57 |
+
"step": 40
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.0,
|
| 61 |
+
"eval_accuracy": 0.8041960784313725,
|
| 62 |
+
"eval_loss": 0.4266883432865143,
|
| 63 |
+
"eval_runtime": 2.8734,
|
| 64 |
+
"eval_samples_per_second": 34.803,
|
| 65 |
+
"eval_steps_per_second": 2.436,
|
| 66 |
+
"step": 44
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.0,
|
| 70 |
+
"eval_exact_match": 12.0,
|
| 71 |
+
"eval_f1": 15.471428571428568,
|
| 72 |
+
"eval_qa_bleu": 6.1719618201256194,
|
| 73 |
+
"eval_qa_exact_match": 0.1,
|
| 74 |
+
"eval_recite_bleu": 20.98411907438438,
|
| 75 |
+
"eval_recite_exact_match": 0.01,
|
| 76 |
+
"step": 44
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 1.02,
|
| 80 |
+
"learning_rate": 3e-05,
|
| 81 |
+
"loss": 0.1653,
|
| 82 |
+
"step": 45
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 1.14,
|
| 86 |
+
"learning_rate": 3e-05,
|
| 87 |
+
"loss": 0.1162,
|
| 88 |
+
"step": 50
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.25,
|
| 92 |
+
"learning_rate": 3e-05,
|
| 93 |
+
"loss": 0.1258,
|
| 94 |
+
"step": 55
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.36,
|
| 98 |
+
"learning_rate": 3e-05,
|
| 99 |
+
"loss": 0.1233,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 1.48,
|
| 104 |
+
"learning_rate": 3e-05,
|
| 105 |
+
"loss": 0.1084,
|
| 106 |
+
"step": 65
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.59,
|
| 110 |
+
"learning_rate": 3e-05,
|
| 111 |
+
"loss": 0.1216,
|
| 112 |
+
"step": 70
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.7,
|
| 116 |
+
"learning_rate": 3e-05,
|
| 117 |
+
"loss": 0.1178,
|
| 118 |
+
"step": 75
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 1.82,
|
| 122 |
+
"learning_rate": 3e-05,
|
| 123 |
+
"loss": 0.1194,
|
| 124 |
+
"step": 80
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 1.93,
|
| 128 |
+
"learning_rate": 3e-05,
|
| 129 |
+
"loss": 0.1121,
|
| 130 |
+
"step": 85
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 2.0,
|
| 134 |
+
"eval_accuracy": 0.813843137254902,
|
| 135 |
+
"eval_loss": 0.37441933155059814,
|
| 136 |
+
"eval_runtime": 2.8664,
|
| 137 |
+
"eval_samples_per_second": 34.887,
|
| 138 |
+
"eval_steps_per_second": 2.442,
|
| 139 |
+
"step": 88
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 2.0,
|
| 143 |
+
"eval_exact_match": 11.0,
|
| 144 |
+
"eval_f1": 13.566666666666665,
|
| 145 |
+
"eval_qa_bleu": 8.02710682428091,
|
| 146 |
+
"eval_qa_exact_match": 0.08,
|
| 147 |
+
"eval_recite_bleu": 20.340555459484488,
|
| 148 |
+
"eval_recite_exact_match": 0.04,
|
| 149 |
+
"step": 88
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 2.05,
|
| 153 |
+
"learning_rate": 3e-05,
|
| 154 |
+
"loss": 0.0929,
|
| 155 |
+
"step": 90
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 2.16,
|
| 159 |
+
"learning_rate": 3e-05,
|
| 160 |
+
"loss": 0.0821,
|
| 161 |
+
"step": 95
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 2.27,
|
| 165 |
+
"learning_rate": 3e-05,
|
| 166 |
+
"loss": 0.1454,
|
| 167 |
+
"step": 100
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.39,
|
| 171 |
+
"learning_rate": 3e-05,
|
| 172 |
+
"loss": 0.0886,
|
| 173 |
+
"step": 105
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 2.5,
|
| 177 |
+
"learning_rate": 3e-05,
|
| 178 |
+
"loss": 0.0872,
|
| 179 |
+
"step": 110
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.61,
|
| 183 |
+
"learning_rate": 3e-05,
|
| 184 |
+
"loss": 0.0763,
|
| 185 |
+
"step": 115
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 2.73,
|
| 189 |
+
"learning_rate": 3e-05,
|
| 190 |
+
"loss": 0.0822,
|
| 191 |
+
"step": 120
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 2.84,
|
| 195 |
+
"learning_rate": 3e-05,
|
| 196 |
+
"loss": 0.0839,
|
| 197 |
+
"step": 125
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 2.95,
|
| 201 |
+
"learning_rate": 3e-05,
|
| 202 |
+
"loss": 0.0832,
|
| 203 |
+
"step": 130
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 3.0,
|
| 207 |
+
"eval_accuracy": 0.8176862745098039,
|
| 208 |
+
"eval_loss": 0.35512271523475647,
|
| 209 |
+
"eval_runtime": 2.9588,
|
| 210 |
+
"eval_samples_per_second": 33.798,
|
| 211 |
+
"eval_steps_per_second": 2.366,
|
| 212 |
+
"step": 132
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 3.0,
|
| 216 |
+
"eval_exact_match": 12.0,
|
| 217 |
+
"eval_f1": 15.433333333333335,
|
| 218 |
+
"eval_qa_bleu": 10.029035778799281,
|
| 219 |
+
"eval_qa_exact_match": 0.09,
|
| 220 |
+
"eval_recite_bleu": 20.92406668269511,
|
| 221 |
+
"eval_recite_exact_match": 0.06,
|
| 222 |
+
"step": 132
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.07,
|
| 226 |
+
"learning_rate": 3e-05,
|
| 227 |
+
"loss": 0.0745,
|
| 228 |
+
"step": 135
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.18,
|
| 232 |
+
"learning_rate": 3e-05,
|
| 233 |
+
"loss": 0.0668,
|
| 234 |
+
"step": 140
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 3.3,
|
| 238 |
+
"learning_rate": 3e-05,
|
| 239 |
+
"loss": 0.0686,
|
| 240 |
+
"step": 145
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.41,
|
| 244 |
+
"learning_rate": 3e-05,
|
| 245 |
+
"loss": 0.0764,
|
| 246 |
+
"step": 150
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.52,
|
| 250 |
+
"learning_rate": 3e-05,
|
| 251 |
+
"loss": 0.0678,
|
| 252 |
+
"step": 155
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 3.64,
|
| 256 |
+
"learning_rate": 3e-05,
|
| 257 |
+
"loss": 0.0667,
|
| 258 |
+
"step": 160
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 3.75,
|
| 262 |
+
"learning_rate": 3e-05,
|
| 263 |
+
"loss": 0.0683,
|
| 264 |
+
"step": 165
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.86,
|
| 268 |
+
"learning_rate": 3e-05,
|
| 269 |
+
"loss": 0.0702,
|
| 270 |
+
"step": 170
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 3.98,
|
| 274 |
+
"learning_rate": 3e-05,
|
| 275 |
+
"loss": 0.0859,
|
| 276 |
+
"step": 175
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 4.0,
|
| 280 |
+
"eval_accuracy": 0.8186666666666667,
|
| 281 |
+
"eval_loss": 0.3487281799316406,
|
| 282 |
+
"eval_runtime": 2.9553,
|
| 283 |
+
"eval_samples_per_second": 33.837,
|
| 284 |
+
"eval_steps_per_second": 2.369,
|
| 285 |
+
"step": 176
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.0,
|
| 289 |
+
"eval_exact_match": 11.0,
|
| 290 |
+
"eval_f1": 15.204761904761906,
|
| 291 |
+
"eval_qa_bleu": 6.940011634600674,
|
| 292 |
+
"eval_qa_exact_match": 0.09,
|
| 293 |
+
"eval_recite_bleu": 21.779007247778743,
|
| 294 |
+
"eval_recite_exact_match": 0.06,
|
| 295 |
+
"step": 176
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 4.09,
|
| 299 |
+
"learning_rate": 3e-05,
|
| 300 |
+
"loss": 0.0765,
|
| 301 |
+
"step": 180
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 4.2,
|
| 305 |
+
"learning_rate": 3e-05,
|
| 306 |
+
"loss": 0.08,
|
| 307 |
+
"step": 185
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 4.32,
|
| 311 |
+
"learning_rate": 3e-05,
|
| 312 |
+
"loss": 0.0812,
|
| 313 |
+
"step": 190
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 4.43,
|
| 317 |
+
"learning_rate": 3e-05,
|
| 318 |
+
"loss": 0.0836,
|
| 319 |
+
"step": 195
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 4.55,
|
| 323 |
+
"learning_rate": 3e-05,
|
| 324 |
+
"loss": 0.0831,
|
| 325 |
+
"step": 200
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 4.66,
|
| 329 |
+
"learning_rate": 3e-05,
|
| 330 |
+
"loss": 0.0849,
|
| 331 |
+
"step": 205
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 4.77,
|
| 335 |
+
"learning_rate": 3e-05,
|
| 336 |
+
"loss": 0.0785,
|
| 337 |
+
"step": 210
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 4.89,
|
| 341 |
+
"learning_rate": 3e-05,
|
| 342 |
+
"loss": 0.0859,
|
| 343 |
+
"step": 215
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 5.0,
|
| 347 |
+
"learning_rate": 3e-05,
|
| 348 |
+
"loss": 0.0849,
|
| 349 |
+
"step": 220
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 5.0,
|
| 353 |
+
"eval_accuracy": 0.8187843137254902,
|
| 354 |
+
"eval_loss": 0.3478342294692993,
|
| 355 |
+
"eval_runtime": 2.8659,
|
| 356 |
+
"eval_samples_per_second": 34.894,
|
| 357 |
+
"eval_steps_per_second": 2.443,
|
| 358 |
+
"step": 220
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 5.0,
|
| 362 |
+
"eval_exact_match": 7.0,
|
| 363 |
+
"eval_f1": 13.27936507936508,
|
| 364 |
+
"eval_qa_bleu": 7.421149129461378,
|
| 365 |
+
"eval_qa_exact_match": 0.05,
|
| 366 |
+
"eval_recite_bleu": 24.669275256593362,
|
| 367 |
+
"eval_recite_exact_match": 0.09,
|
| 368 |
+
"step": 220
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 5.11,
|
| 372 |
+
"learning_rate": 3e-05,
|
| 373 |
+
"loss": 0.0724,
|
| 374 |
+
"step": 225
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 5.23,
|
| 378 |
+
"learning_rate": 3e-05,
|
| 379 |
+
"loss": 0.0758,
|
| 380 |
+
"step": 230
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 5.34,
|
| 384 |
+
"learning_rate": 3e-05,
|
| 385 |
+
"loss": 0.0729,
|
| 386 |
+
"step": 235
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 5.45,
|
| 390 |
+
"learning_rate": 3e-05,
|
| 391 |
+
"loss": 0.0804,
|
| 392 |
+
"step": 240
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 5.57,
|
| 396 |
+
"learning_rate": 3e-05,
|
| 397 |
+
"loss": 0.0739,
|
| 398 |
+
"step": 245
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 5.68,
|
| 402 |
+
"learning_rate": 3e-05,
|
| 403 |
+
"loss": 0.0762,
|
| 404 |
+
"step": 250
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 5.8,
|
| 408 |
+
"learning_rate": 3e-05,
|
| 409 |
+
"loss": 0.0758,
|
| 410 |
+
"step": 255
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 5.91,
|
| 414 |
+
"learning_rate": 3e-05,
|
| 415 |
+
"loss": 0.0791,
|
| 416 |
+
"step": 260
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 6.0,
|
| 420 |
+
"eval_accuracy": 0.8193725490196079,
|
| 421 |
+
"eval_loss": 0.3478553295135498,
|
| 422 |
+
"eval_runtime": 2.944,
|
| 423 |
+
"eval_samples_per_second": 33.968,
|
| 424 |
+
"eval_steps_per_second": 2.378,
|
| 425 |
+
"step": 264
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 6.0,
|
| 429 |
+
"eval_exact_match": 9.0,
|
| 430 |
+
"eval_f1": 12.366666666666667,
|
| 431 |
+
"eval_qa_bleu": 7.573063970329312,
|
| 432 |
+
"eval_qa_exact_match": 0.08,
|
| 433 |
+
"eval_recite_bleu": 21.721229940044044,
|
| 434 |
+
"eval_recite_exact_match": 0.07,
|
| 435 |
+
"step": 264
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 6.02,
|
| 439 |
+
"learning_rate": 3e-05,
|
| 440 |
+
"loss": 0.0742,
|
| 441 |
+
"step": 265
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.14,
|
| 445 |
+
"learning_rate": 3e-05,
|
| 446 |
+
"loss": 0.0611,
|
| 447 |
+
"step": 270
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 6.25,
|
| 451 |
+
"learning_rate": 3e-05,
|
| 452 |
+
"loss": 0.067,
|
| 453 |
+
"step": 275
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 6.36,
|
| 457 |
+
"learning_rate": 3e-05,
|
| 458 |
+
"loss": 0.071,
|
| 459 |
+
"step": 280
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 6.48,
|
| 463 |
+
"learning_rate": 3e-05,
|
| 464 |
+
"loss": 0.0714,
|
| 465 |
+
"step": 285
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 6.59,
|
| 469 |
+
"learning_rate": 3e-05,
|
| 470 |
+
"loss": 0.07,
|
| 471 |
+
"step": 290
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 6.7,
|
| 475 |
+
"learning_rate": 3e-05,
|
| 476 |
+
"loss": 0.0696,
|
| 477 |
+
"step": 295
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 6.82,
|
| 481 |
+
"learning_rate": 3e-05,
|
| 482 |
+
"loss": 0.0753,
|
| 483 |
+
"step": 300
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 6.93,
|
| 487 |
+
"learning_rate": 3e-05,
|
| 488 |
+
"loss": 0.0681,
|
| 489 |
+
"step": 305
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 7.0,
|
| 493 |
+
"eval_accuracy": 0.8196862745098039,
|
| 494 |
+
"eval_loss": 0.35373786091804504,
|
| 495 |
+
"eval_runtime": 2.875,
|
| 496 |
+
"eval_samples_per_second": 34.783,
|
| 497 |
+
"eval_steps_per_second": 2.435,
|
| 498 |
+
"step": 308
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 7.0,
|
| 502 |
+
"eval_exact_match": 11.0,
|
| 503 |
+
"eval_f1": 14.133333333333333,
|
| 504 |
+
"eval_qa_bleu": 10.202817947299275,
|
| 505 |
+
"eval_qa_exact_match": 0.08,
|
| 506 |
+
"eval_recite_bleu": 27.833105068617286,
|
| 507 |
+
"eval_recite_exact_match": 0.1,
|
| 508 |
+
"step": 308
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 7.05,
|
| 512 |
+
"learning_rate": 3e-05,
|
| 513 |
+
"loss": 0.0831,
|
| 514 |
+
"step": 310
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 7.16,
|
| 518 |
+
"learning_rate": 3e-05,
|
| 519 |
+
"loss": 0.0874,
|
| 520 |
+
"step": 315
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.27,
|
| 524 |
+
"learning_rate": 3e-05,
|
| 525 |
+
"loss": 0.0775,
|
| 526 |
+
"step": 320
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 7.39,
|
| 530 |
+
"learning_rate": 3e-05,
|
| 531 |
+
"loss": 0.0715,
|
| 532 |
+
"step": 325
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 7.5,
|
| 536 |
+
"learning_rate": 3e-05,
|
| 537 |
+
"loss": 0.0688,
|
| 538 |
+
"step": 330
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 7.61,
|
| 542 |
+
"learning_rate": 3e-05,
|
| 543 |
+
"loss": 0.0689,
|
| 544 |
+
"step": 335
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 7.73,
|
| 548 |
+
"learning_rate": 3e-05,
|
| 549 |
+
"loss": 0.0695,
|
| 550 |
+
"step": 340
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 7.84,
|
| 554 |
+
"learning_rate": 3e-05,
|
| 555 |
+
"loss": 0.0707,
|
| 556 |
+
"step": 345
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 7.95,
|
| 560 |
+
"learning_rate": 3e-05,
|
| 561 |
+
"loss": 0.0715,
|
| 562 |
+
"step": 350
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 8.0,
|
| 566 |
+
"eval_accuracy": 0.8205098039215686,
|
| 567 |
+
"eval_loss": 0.3466941714286804,
|
| 568 |
+
"eval_runtime": 3.016,
|
| 569 |
+
"eval_samples_per_second": 33.157,
|
| 570 |
+
"eval_steps_per_second": 2.321,
|
| 571 |
+
"step": 352
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 8.0,
|
| 575 |
+
"eval_exact_match": 12.0,
|
| 576 |
+
"eval_f1": 17.03333333333333,
|
| 577 |
+
"eval_qa_bleu": 4.84377651955754,
|
| 578 |
+
"eval_qa_exact_match": 0.1,
|
| 579 |
+
"eval_recite_bleu": 29.309088448185722,
|
| 580 |
+
"eval_recite_exact_match": 0.15,
|
| 581 |
+
"step": 352
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 8.07,
|
| 585 |
+
"learning_rate": 3e-05,
|
| 586 |
+
"loss": 0.0591,
|
| 587 |
+
"step": 355
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 8.18,
|
| 591 |
+
"learning_rate": 3e-05,
|
| 592 |
+
"loss": 0.0735,
|
| 593 |
+
"step": 360
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 8.3,
|
| 597 |
+
"learning_rate": 3e-05,
|
| 598 |
+
"loss": 0.098,
|
| 599 |
+
"step": 365
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 8.41,
|
| 603 |
+
"learning_rate": 3e-05,
|
| 604 |
+
"loss": 0.1085,
|
| 605 |
+
"step": 370
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 8.52,
|
| 609 |
+
"learning_rate": 3e-05,
|
| 610 |
+
"loss": 0.0865,
|
| 611 |
+
"step": 375
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 8.64,
|
| 615 |
+
"learning_rate": 3e-05,
|
| 616 |
+
"loss": 0.0783,
|
| 617 |
+
"step": 380
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 8.75,
|
| 621 |
+
"learning_rate": 3e-05,
|
| 622 |
+
"loss": 0.0749,
|
| 623 |
+
"step": 385
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 8.86,
|
| 627 |
+
"learning_rate": 3e-05,
|
| 628 |
+
"loss": 0.0764,
|
| 629 |
+
"step": 390
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 8.98,
|
| 633 |
+
"learning_rate": 3e-05,
|
| 634 |
+
"loss": 0.0639,
|
| 635 |
+
"step": 395
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 9.0,
|
| 639 |
+
"eval_accuracy": 0.8194117647058824,
|
| 640 |
+
"eval_loss": 0.35771244764328003,
|
| 641 |
+
"eval_runtime": 2.9974,
|
| 642 |
+
"eval_samples_per_second": 33.363,
|
| 643 |
+
"eval_steps_per_second": 2.335,
|
| 644 |
+
"step": 396
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 9.0,
|
| 648 |
+
"eval_exact_match": 10.0,
|
| 649 |
+
"eval_f1": 13.91904761904762,
|
| 650 |
+
"eval_qa_bleu": 4.263983180255635,
|
| 651 |
+
"eval_qa_exact_match": 0.07,
|
| 652 |
+
"eval_recite_bleu": 27.571657075027684,
|
| 653 |
+
"eval_recite_exact_match": 0.11,
|
| 654 |
+
"step": 396
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 9.09,
|
| 658 |
+
"learning_rate": 3e-05,
|
| 659 |
+
"loss": 0.0579,
|
| 660 |
+
"step": 400
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 9.2,
|
| 664 |
+
"learning_rate": 3e-05,
|
| 665 |
+
"loss": 0.0602,
|
| 666 |
+
"step": 405
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 9.32,
|
| 670 |
+
"learning_rate": 3e-05,
|
| 671 |
+
"loss": 0.0585,
|
| 672 |
+
"step": 410
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 9.43,
|
| 676 |
+
"learning_rate": 3e-05,
|
| 677 |
+
"loss": 0.0618,
|
| 678 |
+
"step": 415
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 9.55,
|
| 682 |
+
"learning_rate": 3e-05,
|
| 683 |
+
"loss": 0.0622,
|
| 684 |
+
"step": 420
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 9.66,
|
| 688 |
+
"learning_rate": 3e-05,
|
| 689 |
+
"loss": 0.0606,
|
| 690 |
+
"step": 425
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 9.77,
|
| 694 |
+
"learning_rate": 3e-05,
|
| 695 |
+
"loss": 0.0695,
|
| 696 |
+
"step": 430
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 9.89,
|
| 700 |
+
"learning_rate": 3e-05,
|
| 701 |
+
"loss": 0.0575,
|
| 702 |
+
"step": 435
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 10.0,
|
| 706 |
+
"learning_rate": 3e-05,
|
| 707 |
+
"loss": 0.0614,
|
| 708 |
+
"step": 440
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 10.0,
|
| 712 |
+
"eval_accuracy": 0.8212156862745098,
|
| 713 |
+
"eval_loss": 0.33876052498817444,
|
| 714 |
+
"eval_runtime": 2.9196,
|
| 715 |
+
"eval_samples_per_second": 34.251,
|
| 716 |
+
"eval_steps_per_second": 2.398,
|
| 717 |
+
"step": 440
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 10.0,
|
| 721 |
+
"eval_exact_match": 15.0,
|
| 722 |
+
"eval_f1": 20.53333333333333,
|
| 723 |
+
"eval_qa_bleu": 6.622399363390335,
|
| 724 |
+
"eval_qa_exact_match": 0.13,
|
| 725 |
+
"eval_recite_bleu": 38.34543604737045,
|
| 726 |
+
"eval_recite_exact_match": 0.16,
|
| 727 |
+
"step": 440
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 10.0,
|
| 731 |
+
"step": 440,
|
| 732 |
+
"total_flos": 1.16589670127616e+16,
|
| 733 |
+
"train_loss": 0.09364926754073663,
|
| 734 |
+
"train_runtime": 1254.2575,
|
| 735 |
+
"train_samples_per_second": 5.517,
|
| 736 |
+
"train_steps_per_second": 0.351
|
| 737 |
+
}
|
| 738 |
+
],
|
| 739 |
+
"logging_steps": 5,
|
| 740 |
+
"max_steps": 440,
|
| 741 |
+
"num_train_epochs": 10,
|
| 742 |
+
"save_steps": 500,
|
| 743 |
+
"total_flos": 1.16589670127616e+16,
|
| 744 |
+
"trial_name": null,
|
| 745 |
+
"trial_params": null
|
| 746 |
+
}
|