End of training
Browse files- README.md +11 -7
- all_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +456 -58
README.md
CHANGED
|
@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 20 |
|
| 21 |
This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on an unknown dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
-
- Loss: 1.
|
| 24 |
-
- Accuracy: 0.
|
| 25 |
-
- F1: 0.
|
| 26 |
-
- Precision: 0.
|
| 27 |
-
- Recall: 0.
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
|
@@ -49,13 +49,17 @@ The following hyperparameters were used during training:
|
|
| 49 |
- seed: 42
|
| 50 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
- lr_scheduler_type: linear
|
| 52 |
-
- num_epochs:
|
| 53 |
|
| 54 |
### Training results
|
| 55 |
|
| 56 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|
| 57 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
|
| 58 |
-
| 1.
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
|
| 60 |
|
| 61 |
### Framework versions
|
|
|
|
| 20 |
|
| 21 |
This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on an unknown dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.3012
|
| 24 |
+
- Accuracy: 0.5435
|
| 25 |
+
- F1: 0.5062
|
| 26 |
+
- Precision: 0.5113
|
| 27 |
+
- Recall: 0.5435
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
|
|
|
| 49 |
- seed: 42
|
| 50 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
- lr_scheduler_type: linear
|
| 52 |
+
- num_epochs: 5
|
| 53 |
|
| 54 |
### Training results
|
| 55 |
|
| 56 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|
| 57 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
|
| 58 |
+
| 1.6077 | 1.0 | 125 | 1.6378 | 0.4130 | 0.3268 | 0.2997 | 0.4130 |
|
| 59 |
+
| 1.6016 | 2.0 | 250 | 1.4609 | 0.4870 | 0.4109 | 0.3904 | 0.4870 |
|
| 60 |
+
| 1.2479 | 3.0 | 375 | 1.4185 | 0.5043 | 0.4485 | 0.4236 | 0.5043 |
|
| 61 |
+
| 1.1542 | 4.0 | 500 | 1.3072 | 0.5435 | 0.5141 | 0.5397 | 0.5435 |
|
| 62 |
+
| 1.1302 | 5.0 | 625 | 1.3012 | 0.5435 | 0.5062 | 0.5113 | 0.5435 |
|
| 63 |
|
| 64 |
|
| 65 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 1.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second":
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 1.4087707061767578,
|
| 4 |
+
"train_runtime": 222.2422,
|
| 5 |
+
"train_samples_per_second": 44.906,
|
| 6 |
+
"train_steps_per_second": 2.812
|
| 7 |
}
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 1.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second":
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 1.4087707061767578,
|
| 4 |
+
"train_runtime": 222.2422,
|
| 5 |
+
"train_samples_per_second": 44.906,
|
| 6 |
+
"train_steps_per_second": 2.812
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,125 +1,523 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 1.
|
| 3 |
-
"best_model_checkpoint": "roberta-base-CD_baseline/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.08,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 1.
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.16,
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate": 1.
|
| 22 |
-
"loss": 2.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.24,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 1.
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.32,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 1.
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.4,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 1.
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.48,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate": 1.
|
| 50 |
-
"loss": 1.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.56,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 1.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.64,
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 1.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.72,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.8,
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 1.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.88,
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 1.
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.96,
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 1.
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 1.0,
|
| 97 |
-
"eval_accuracy": 0.
|
| 98 |
-
"eval_f1": 0.
|
| 99 |
-
"eval_loss": 1.
|
| 100 |
-
"eval_precision": 0.
|
| 101 |
-
"eval_recall": 0.
|
| 102 |
-
"eval_runtime": 1.
|
| 103 |
-
"eval_samples_per_second": 176.
|
| 104 |
-
"eval_steps_per_second": 11.
|
| 105 |
"step": 125
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch": 1.
|
| 109 |
-
"
|
| 110 |
-
"
|
| 111 |
-
"
|
| 112 |
-
"
|
| 113 |
-
|
| 114 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 115 |
}
|
| 116 |
],
|
| 117 |
"logging_steps": 10,
|
| 118 |
-
"max_steps":
|
| 119 |
"num_input_tokens_seen": 0,
|
| 120 |
-
"num_train_epochs":
|
| 121 |
"save_steps": 500,
|
| 122 |
-
"total_flos":
|
| 123 |
"train_batch_size": 16,
|
| 124 |
"trial_name": null,
|
| 125 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 1.301220178604126,
|
| 3 |
+
"best_model_checkpoint": "roberta-base-CD_baseline/checkpoint-625",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 625,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.08,
|
| 13 |
+
"grad_norm": 4.637064456939697,
|
| 14 |
+
"learning_rate": 1.968e-05,
|
| 15 |
+
"loss": 2.2753,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.16,
|
| 20 |
+
"grad_norm": 15.68393325805664,
|
| 21 |
+
"learning_rate": 1.936e-05,
|
| 22 |
+
"loss": 2.0418,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.24,
|
| 27 |
+
"grad_norm": 12.480018615722656,
|
| 28 |
+
"learning_rate": 1.904e-05,
|
| 29 |
+
"loss": 1.9195,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.32,
|
| 34 |
+
"grad_norm": 41.00396728515625,
|
| 35 |
+
"learning_rate": 1.8720000000000004e-05,
|
| 36 |
+
"loss": 1.9629,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.4,
|
| 41 |
+
"grad_norm": 4.204047203063965,
|
| 42 |
+
"learning_rate": 1.8400000000000003e-05,
|
| 43 |
+
"loss": 1.7429,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.48,
|
| 48 |
+
"grad_norm": 8.693258285522461,
|
| 49 |
+
"learning_rate": 1.8080000000000003e-05,
|
| 50 |
+
"loss": 1.9048,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.56,
|
| 55 |
+
"grad_norm": 3.7029831409454346,
|
| 56 |
+
"learning_rate": 1.7760000000000003e-05,
|
| 57 |
+
"loss": 1.8083,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.64,
|
| 62 |
+
"grad_norm": 4.148443698883057,
|
| 63 |
+
"learning_rate": 1.7440000000000002e-05,
|
| 64 |
+
"loss": 1.6655,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.72,
|
| 69 |
+
"grad_norm": 20.234933853149414,
|
| 70 |
+
"learning_rate": 1.7120000000000002e-05,
|
| 71 |
+
"loss": 1.5108,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.8,
|
| 76 |
+
"grad_norm": 8.840304374694824,
|
| 77 |
+
"learning_rate": 1.6800000000000002e-05,
|
| 78 |
+
"loss": 1.7,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.88,
|
| 83 |
+
"grad_norm": 7.2302350997924805,
|
| 84 |
+
"learning_rate": 1.648e-05,
|
| 85 |
+
"loss": 1.6971,
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.96,
|
| 90 |
+
"grad_norm": 3.1535422801971436,
|
| 91 |
+
"learning_rate": 1.616e-05,
|
| 92 |
+
"loss": 1.6077,
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 1.0,
|
| 97 |
+
"eval_accuracy": 0.41304347826086957,
|
| 98 |
+
"eval_f1": 0.3267868906455863,
|
| 99 |
+
"eval_loss": 1.6377873420715332,
|
| 100 |
+
"eval_precision": 0.29971778816474115,
|
| 101 |
+
"eval_recall": 0.41304347826086957,
|
| 102 |
+
"eval_runtime": 1.3051,
|
| 103 |
+
"eval_samples_per_second": 176.233,
|
| 104 |
+
"eval_steps_per_second": 11.493,
|
| 105 |
"step": 125
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"epoch": 1.04,
|
| 109 |
+
"grad_norm": 9.82644271850586,
|
| 110 |
+
"learning_rate": 1.584e-05,
|
| 111 |
+
"loss": 1.7461,
|
| 112 |
+
"step": 130
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.12,
|
| 116 |
+
"grad_norm": 7.467548370361328,
|
| 117 |
+
"learning_rate": 1.552e-05,
|
| 118 |
+
"loss": 1.6311,
|
| 119 |
+
"step": 140
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.2,
|
| 123 |
+
"grad_norm": 5.879318714141846,
|
| 124 |
+
"learning_rate": 1.5200000000000002e-05,
|
| 125 |
+
"loss": 1.4653,
|
| 126 |
+
"step": 150
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.28,
|
| 130 |
+
"grad_norm": 12.443910598754883,
|
| 131 |
+
"learning_rate": 1.4880000000000002e-05,
|
| 132 |
+
"loss": 1.628,
|
| 133 |
+
"step": 160
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 1.36,
|
| 137 |
+
"grad_norm": 5.35117244720459,
|
| 138 |
+
"learning_rate": 1.4560000000000001e-05,
|
| 139 |
+
"loss": 1.5741,
|
| 140 |
+
"step": 170
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.44,
|
| 144 |
+
"grad_norm": 7.006358623504639,
|
| 145 |
+
"learning_rate": 1.4240000000000001e-05,
|
| 146 |
+
"loss": 1.8264,
|
| 147 |
+
"step": 180
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.52,
|
| 151 |
+
"grad_norm": 5.610796928405762,
|
| 152 |
+
"learning_rate": 1.392e-05,
|
| 153 |
+
"loss": 1.5304,
|
| 154 |
+
"step": 190
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.6,
|
| 158 |
+
"grad_norm": 23.96417808532715,
|
| 159 |
+
"learning_rate": 1.3600000000000002e-05,
|
| 160 |
+
"loss": 1.4833,
|
| 161 |
+
"step": 200
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.68,
|
| 165 |
+
"grad_norm": 6.071532249450684,
|
| 166 |
+
"learning_rate": 1.3280000000000002e-05,
|
| 167 |
+
"loss": 1.4721,
|
| 168 |
+
"step": 210
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.76,
|
| 172 |
+
"grad_norm": 27.290363311767578,
|
| 173 |
+
"learning_rate": 1.2960000000000001e-05,
|
| 174 |
+
"loss": 1.6517,
|
| 175 |
+
"step": 220
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 1.84,
|
| 179 |
+
"grad_norm": 7.392124176025391,
|
| 180 |
+
"learning_rate": 1.2640000000000001e-05,
|
| 181 |
+
"loss": 1.5183,
|
| 182 |
+
"step": 230
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.92,
|
| 186 |
+
"grad_norm": 8.155671119689941,
|
| 187 |
+
"learning_rate": 1.232e-05,
|
| 188 |
+
"loss": 1.4147,
|
| 189 |
+
"step": 240
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.0,
|
| 193 |
+
"grad_norm": 7.289750099182129,
|
| 194 |
+
"learning_rate": 1.2e-05,
|
| 195 |
+
"loss": 1.6016,
|
| 196 |
+
"step": 250
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 2.0,
|
| 200 |
+
"eval_accuracy": 0.48695652173913045,
|
| 201 |
+
"eval_f1": 0.4109191960797593,
|
| 202 |
+
"eval_loss": 1.460945963859558,
|
| 203 |
+
"eval_precision": 0.39039609511101225,
|
| 204 |
+
"eval_recall": 0.48695652173913045,
|
| 205 |
+
"eval_runtime": 1.2886,
|
| 206 |
+
"eval_samples_per_second": 178.493,
|
| 207 |
+
"eval_steps_per_second": 11.641,
|
| 208 |
+
"step": 250
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.08,
|
| 212 |
+
"grad_norm": 15.545705795288086,
|
| 213 |
+
"learning_rate": 1.168e-05,
|
| 214 |
+
"loss": 1.3776,
|
| 215 |
+
"step": 260
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 2.16,
|
| 219 |
+
"grad_norm": 7.504268646240234,
|
| 220 |
+
"learning_rate": 1.136e-05,
|
| 221 |
+
"loss": 1.3455,
|
| 222 |
+
"step": 270
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.24,
|
| 226 |
+
"grad_norm": 8.06009292602539,
|
| 227 |
+
"learning_rate": 1.1040000000000001e-05,
|
| 228 |
+
"loss": 1.4793,
|
| 229 |
+
"step": 280
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 2.32,
|
| 233 |
+
"grad_norm": 18.173667907714844,
|
| 234 |
+
"learning_rate": 1.072e-05,
|
| 235 |
+
"loss": 1.4226,
|
| 236 |
+
"step": 290
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.4,
|
| 240 |
+
"grad_norm": 11.790044784545898,
|
| 241 |
+
"learning_rate": 1.04e-05,
|
| 242 |
+
"loss": 1.386,
|
| 243 |
+
"step": 300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.48,
|
| 247 |
+
"grad_norm": 7.876099109649658,
|
| 248 |
+
"learning_rate": 1.008e-05,
|
| 249 |
+
"loss": 1.3522,
|
| 250 |
+
"step": 310
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.56,
|
| 254 |
+
"grad_norm": 10.516275405883789,
|
| 255 |
+
"learning_rate": 9.760000000000001e-06,
|
| 256 |
+
"loss": 1.4296,
|
| 257 |
+
"step": 320
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 2.64,
|
| 261 |
+
"grad_norm": 8.990633010864258,
|
| 262 |
+
"learning_rate": 9.440000000000001e-06,
|
| 263 |
+
"loss": 1.4941,
|
| 264 |
+
"step": 330
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 2.72,
|
| 268 |
+
"grad_norm": 8.365500450134277,
|
| 269 |
+
"learning_rate": 9.12e-06,
|
| 270 |
+
"loss": 1.4153,
|
| 271 |
+
"step": 340
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 2.8,
|
| 275 |
+
"grad_norm": 9.450441360473633,
|
| 276 |
+
"learning_rate": 8.8e-06,
|
| 277 |
+
"loss": 1.522,
|
| 278 |
+
"step": 350
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.88,
|
| 282 |
+
"grad_norm": 18.165035247802734,
|
| 283 |
+
"learning_rate": 8.48e-06,
|
| 284 |
+
"loss": 1.3895,
|
| 285 |
+
"step": 360
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.96,
|
| 289 |
+
"grad_norm": 10.70251750946045,
|
| 290 |
+
"learning_rate": 8.16e-06,
|
| 291 |
+
"loss": 1.2479,
|
| 292 |
+
"step": 370
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 3.0,
|
| 296 |
+
"eval_accuracy": 0.5043478260869565,
|
| 297 |
+
"eval_f1": 0.44849074152912793,
|
| 298 |
+
"eval_loss": 1.4184553623199463,
|
| 299 |
+
"eval_precision": 0.4235901359719885,
|
| 300 |
+
"eval_recall": 0.5043478260869565,
|
| 301 |
+
"eval_runtime": 1.303,
|
| 302 |
+
"eval_samples_per_second": 176.51,
|
| 303 |
+
"eval_steps_per_second": 11.511,
|
| 304 |
+
"step": 375
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 3.04,
|
| 308 |
+
"grad_norm": 23.491443634033203,
|
| 309 |
+
"learning_rate": 7.840000000000001e-06,
|
| 310 |
+
"loss": 1.2631,
|
| 311 |
+
"step": 380
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 3.12,
|
| 315 |
+
"grad_norm": 8.99567985534668,
|
| 316 |
+
"learning_rate": 7.520000000000001e-06,
|
| 317 |
+
"loss": 1.0625,
|
| 318 |
+
"step": 390
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 3.2,
|
| 322 |
+
"grad_norm": 12.7035493850708,
|
| 323 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 324 |
+
"loss": 1.3823,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 3.28,
|
| 329 |
+
"grad_norm": 15.28470230102539,
|
| 330 |
+
"learning_rate": 6.88e-06,
|
| 331 |
+
"loss": 1.2633,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 3.36,
|
| 336 |
+
"grad_norm": 10.78918170928955,
|
| 337 |
+
"learning_rate": 6.560000000000001e-06,
|
| 338 |
+
"loss": 1.1007,
|
| 339 |
+
"step": 420
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.44,
|
| 343 |
+
"grad_norm": 14.376765251159668,
|
| 344 |
+
"learning_rate": 6.24e-06,
|
| 345 |
+
"loss": 1.2674,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 3.52,
|
| 350 |
+
"grad_norm": 11.458234786987305,
|
| 351 |
+
"learning_rate": 5.92e-06,
|
| 352 |
+
"loss": 1.0922,
|
| 353 |
+
"step": 440
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 3.6,
|
| 357 |
+
"grad_norm": 15.360867500305176,
|
| 358 |
+
"learning_rate": 5.600000000000001e-06,
|
| 359 |
+
"loss": 1.2077,
|
| 360 |
+
"step": 450
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 3.68,
|
| 364 |
+
"grad_norm": 20.423187255859375,
|
| 365 |
+
"learning_rate": 5.28e-06,
|
| 366 |
+
"loss": 1.302,
|
| 367 |
+
"step": 460
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 3.76,
|
| 371 |
+
"grad_norm": 13.024535179138184,
|
| 372 |
+
"learning_rate": 4.960000000000001e-06,
|
| 373 |
+
"loss": 1.1218,
|
| 374 |
+
"step": 470
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 3.84,
|
| 378 |
+
"grad_norm": 14.865490913391113,
|
| 379 |
+
"learning_rate": 4.6400000000000005e-06,
|
| 380 |
+
"loss": 1.1858,
|
| 381 |
+
"step": 480
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.92,
|
| 385 |
+
"grad_norm": 17.874656677246094,
|
| 386 |
+
"learning_rate": 4.32e-06,
|
| 387 |
+
"loss": 1.1319,
|
| 388 |
+
"step": 490
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 4.0,
|
| 392 |
+
"grad_norm": 16.873672485351562,
|
| 393 |
+
"learning_rate": 4.000000000000001e-06,
|
| 394 |
+
"loss": 1.1542,
|
| 395 |
+
"step": 500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 4.0,
|
| 399 |
+
"eval_accuracy": 0.5434782608695652,
|
| 400 |
+
"eval_f1": 0.5141449277787993,
|
| 401 |
+
"eval_loss": 1.3071645498275757,
|
| 402 |
+
"eval_precision": 0.5397252083665127,
|
| 403 |
+
"eval_recall": 0.5434782608695652,
|
| 404 |
+
"eval_runtime": 2.7726,
|
| 405 |
+
"eval_samples_per_second": 82.954,
|
| 406 |
+
"eval_steps_per_second": 5.41,
|
| 407 |
+
"step": 500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 4.08,
|
| 411 |
+
"grad_norm": 16.543500900268555,
|
| 412 |
+
"learning_rate": 3.6800000000000003e-06,
|
| 413 |
+
"loss": 1.0007,
|
| 414 |
+
"step": 510
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.16,
|
| 418 |
+
"grad_norm": 13.177757263183594,
|
| 419 |
+
"learning_rate": 3.3600000000000004e-06,
|
| 420 |
+
"loss": 1.0294,
|
| 421 |
+
"step": 520
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 4.24,
|
| 425 |
+
"grad_norm": 10.053601264953613,
|
| 426 |
+
"learning_rate": 3.04e-06,
|
| 427 |
+
"loss": 1.1163,
|
| 428 |
+
"step": 530
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.32,
|
| 432 |
+
"grad_norm": 20.3472957611084,
|
| 433 |
+
"learning_rate": 2.7200000000000002e-06,
|
| 434 |
+
"loss": 1.1222,
|
| 435 |
+
"step": 540
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.4,
|
| 439 |
+
"grad_norm": 10.218711853027344,
|
| 440 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 441 |
+
"loss": 0.964,
|
| 442 |
+
"step": 550
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 4.48,
|
| 446 |
+
"grad_norm": 10.360238075256348,
|
| 447 |
+
"learning_rate": 2.08e-06,
|
| 448 |
+
"loss": 1.0094,
|
| 449 |
+
"step": 560
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 4.56,
|
| 453 |
+
"grad_norm": 16.11949348449707,
|
| 454 |
+
"learning_rate": 1.76e-06,
|
| 455 |
+
"loss": 1.0102,
|
| 456 |
+
"step": 570
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 4.64,
|
| 460 |
+
"grad_norm": 14.713994979858398,
|
| 461 |
+
"learning_rate": 1.44e-06,
|
| 462 |
+
"loss": 1.0562,
|
| 463 |
+
"step": 580
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 4.72,
|
| 467 |
+
"grad_norm": 12.219269752502441,
|
| 468 |
+
"learning_rate": 1.12e-06,
|
| 469 |
+
"loss": 0.9808,
|
| 470 |
+
"step": 590
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.8,
|
| 474 |
+
"grad_norm": 27.00701141357422,
|
| 475 |
+
"learning_rate": 8.000000000000001e-07,
|
| 476 |
+
"loss": 1.1036,
|
| 477 |
+
"step": 600
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 4.88,
|
| 481 |
+
"grad_norm": 18.196247100830078,
|
| 482 |
+
"learning_rate": 4.800000000000001e-07,
|
| 483 |
+
"loss": 1.1907,
|
| 484 |
+
"step": 610
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 4.96,
|
| 488 |
+
"grad_norm": 18.10931968688965,
|
| 489 |
+
"learning_rate": 1.6e-07,
|
| 490 |
+
"loss": 1.1302,
|
| 491 |
+
"step": 620
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 5.0,
|
| 495 |
+
"eval_accuracy": 0.5434782608695652,
|
| 496 |
+
"eval_f1": 0.5062253924435693,
|
| 497 |
+
"eval_loss": 1.301220178604126,
|
| 498 |
+
"eval_precision": 0.511256656273167,
|
| 499 |
+
"eval_recall": 0.5434782608695652,
|
| 500 |
+
"eval_runtime": 2.7886,
|
| 501 |
+
"eval_samples_per_second": 82.478,
|
| 502 |
+
"eval_steps_per_second": 5.379,
|
| 503 |
+
"step": 625
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 5.0,
|
| 507 |
+
"step": 625,
|
| 508 |
+
"total_flos": 1434186246250944.0,
|
| 509 |
+
"train_loss": 1.4087707061767578,
|
| 510 |
+
"train_runtime": 222.2422,
|
| 511 |
+
"train_samples_per_second": 44.906,
|
| 512 |
+
"train_steps_per_second": 2.812
|
| 513 |
}
|
| 514 |
],
|
| 515 |
"logging_steps": 10,
|
| 516 |
+
"max_steps": 625,
|
| 517 |
"num_input_tokens_seen": 0,
|
| 518 |
+
"num_train_epochs": 5,
|
| 519 |
"save_steps": 500,
|
| 520 |
+
"total_flos": 1434186246250944.0,
|
| 521 |
"train_batch_size": 16,
|
| 522 |
"trial_name": null,
|
| 523 |
"trial_params": null
|