End of training
Browse files- README.md +70 -0
- all_results.json +7 -0
- train_results.json +7 -0
- trainer_state.json +524 -0
README.md
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
base_model: mental/mental-roberta-base
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- accuracy
|
| 8 |
+
- f1
|
| 9 |
+
- precision
|
| 10 |
+
- recall
|
| 11 |
+
model-index:
|
| 12 |
+
- name: mental-roberta-base-CD_baseline
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# mental-roberta-base-CD_baseline
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [mental/mental-roberta-base](https://huggingface.co/mental/mental-roberta-base) on an unknown dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.2696
|
| 24 |
+
- Accuracy: 0.5565
|
| 25 |
+
- F1: 0.5303
|
| 26 |
+
- Precision: 0.5330
|
| 27 |
+
- Recall: 0.5565
|
| 28 |
+
|
| 29 |
+
## Model description
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Intended uses & limitations
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training and evaluation data
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Training procedure
|
| 42 |
+
|
| 43 |
+
### Training hyperparameters
|
| 44 |
+
|
| 45 |
+
The following hyperparameters were used during training:
|
| 46 |
+
- learning_rate: 2e-05
|
| 47 |
+
- train_batch_size: 16
|
| 48 |
+
- eval_batch_size: 16
|
| 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 |
+
| 1.5964 | 1.0 | 125 | 1.6031 | 0.4043 | 0.3393 | 0.3328 | 0.4043 |
|
| 59 |
+
| 1.5226 | 2.0 | 250 | 1.4421 | 0.4739 | 0.4077 | 0.3895 | 0.4739 |
|
| 60 |
+
| 1.1656 | 3.0 | 375 | 1.3132 | 0.5261 | 0.4795 | 0.4490 | 0.5261 |
|
| 61 |
+
| 1.1095 | 4.0 | 500 | 1.2819 | 0.5565 | 0.5231 | 0.5156 | 0.5565 |
|
| 62 |
+
| 1.0974 | 5.0 | 625 | 1.2696 | 0.5565 | 0.5303 | 0.5330 | 0.5565 |
|
| 63 |
+
|
| 64 |
+
|
| 65 |
+
### Framework versions
|
| 66 |
+
|
| 67 |
+
- Transformers 4.38.0
|
| 68 |
+
- Pytorch 2.8.0+cu128
|
| 69 |
+
- Datasets 4.2.0
|
| 70 |
+
- Tokenizers 0.15.2
|
all_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 1.3634769760131835,
|
| 4 |
+
"train_runtime": 225.3467,
|
| 5 |
+
"train_samples_per_second": 44.287,
|
| 6 |
+
"train_steps_per_second": 2.774
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 1.3634769760131835,
|
| 4 |
+
"train_runtime": 225.3467,
|
| 5 |
+
"train_samples_per_second": 44.287,
|
| 6 |
+
"train_steps_per_second": 2.774
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,524 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.269640326499939,
|
| 3 |
+
"best_model_checkpoint": "mental-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": 2.921304702758789,
|
| 14 |
+
"learning_rate": 1.968e-05,
|
| 15 |
+
"loss": 2.3689,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.16,
|
| 20 |
+
"grad_norm": 5.469061851501465,
|
| 21 |
+
"learning_rate": 1.936e-05,
|
| 22 |
+
"loss": 2.0549,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.24,
|
| 27 |
+
"grad_norm": 5.983363628387451,
|
| 28 |
+
"learning_rate": 1.904e-05,
|
| 29 |
+
"loss": 1.9011,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.32,
|
| 34 |
+
"grad_norm": 3.8824734687805176,
|
| 35 |
+
"learning_rate": 1.8720000000000004e-05,
|
| 36 |
+
"loss": 1.9096,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.4,
|
| 41 |
+
"grad_norm": 18.308826446533203,
|
| 42 |
+
"learning_rate": 1.8400000000000003e-05,
|
| 43 |
+
"loss": 1.669,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.48,
|
| 48 |
+
"grad_norm": 5.767693996429443,
|
| 49 |
+
"learning_rate": 1.8080000000000003e-05,
|
| 50 |
+
"loss": 1.8629,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.56,
|
| 55 |
+
"grad_norm": 2.5474987030029297,
|
| 56 |
+
"learning_rate": 1.7760000000000003e-05,
|
| 57 |
+
"loss": 1.7754,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.64,
|
| 62 |
+
"grad_norm": 3.5733895301818848,
|
| 63 |
+
"learning_rate": 1.7440000000000002e-05,
|
| 64 |
+
"loss": 1.7045,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.72,
|
| 69 |
+
"grad_norm": 3.388791799545288,
|
| 70 |
+
"learning_rate": 1.7120000000000002e-05,
|
| 71 |
+
"loss": 1.5388,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.8,
|
| 76 |
+
"grad_norm": 2.2617416381835938,
|
| 77 |
+
"learning_rate": 1.6800000000000002e-05,
|
| 78 |
+
"loss": 1.6174,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.88,
|
| 83 |
+
"grad_norm": 2.040916681289673,
|
| 84 |
+
"learning_rate": 1.648e-05,
|
| 85 |
+
"loss": 1.7345,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.96,
|
| 90 |
+
"grad_norm": 1.7220678329467773,
|
| 91 |
+
"learning_rate": 1.616e-05,
|
| 92 |
+
"loss": 1.5964,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"eval_accuracy": 0.4043478260869565,
|
| 98 |
+
"eval_f1": 0.33928429317982867,
|
| 99 |
+
"eval_loss": 1.603085994720459,
|
| 100 |
+
"eval_precision": 0.33282864804603934,
|
| 101 |
+
"eval_recall": 0.4043478260869565,
|
| 102 |
+
"eval_runtime": 1.7794,
|
| 103 |
+
"eval_samples_per_second": 129.257,
|
| 104 |
+
"eval_steps_per_second": 8.43,
|
| 105 |
+
"step": 125
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.04,
|
| 109 |
+
"grad_norm": 15.428503036499023,
|
| 110 |
+
"learning_rate": 1.584e-05,
|
| 111 |
+
"loss": 1.7193,
|
| 112 |
+
"step": 130
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.12,
|
| 116 |
+
"grad_norm": 2.470576524734497,
|
| 117 |
+
"learning_rate": 1.552e-05,
|
| 118 |
+
"loss": 1.5825,
|
| 119 |
+
"step": 140
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.2,
|
| 123 |
+
"grad_norm": 3.9197633266448975,
|
| 124 |
+
"learning_rate": 1.5200000000000002e-05,
|
| 125 |
+
"loss": 1.4732,
|
| 126 |
+
"step": 150
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.28,
|
| 130 |
+
"grad_norm": 6.2479400634765625,
|
| 131 |
+
"learning_rate": 1.4880000000000002e-05,
|
| 132 |
+
"loss": 1.6381,
|
| 133 |
+
"step": 160
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 1.36,
|
| 137 |
+
"grad_norm": 4.600492000579834,
|
| 138 |
+
"learning_rate": 1.4560000000000001e-05,
|
| 139 |
+
"loss": 1.5587,
|
| 140 |
+
"step": 170
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.44,
|
| 144 |
+
"grad_norm": 6.798262596130371,
|
| 145 |
+
"learning_rate": 1.4240000000000001e-05,
|
| 146 |
+
"loss": 1.8529,
|
| 147 |
+
"step": 180
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.52,
|
| 151 |
+
"grad_norm": 4.2226996421813965,
|
| 152 |
+
"learning_rate": 1.392e-05,
|
| 153 |
+
"loss": 1.4755,
|
| 154 |
+
"step": 190
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.6,
|
| 158 |
+
"grad_norm": 5.053036212921143,
|
| 159 |
+
"learning_rate": 1.3600000000000002e-05,
|
| 160 |
+
"loss": 1.4778,
|
| 161 |
+
"step": 200
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.68,
|
| 165 |
+
"grad_norm": 4.096879959106445,
|
| 166 |
+
"learning_rate": 1.3280000000000002e-05,
|
| 167 |
+
"loss": 1.4701,
|
| 168 |
+
"step": 210
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.76,
|
| 172 |
+
"grad_norm": 18.876264572143555,
|
| 173 |
+
"learning_rate": 1.2960000000000001e-05,
|
| 174 |
+
"loss": 1.6647,
|
| 175 |
+
"step": 220
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 1.84,
|
| 179 |
+
"grad_norm": 5.195082664489746,
|
| 180 |
+
"learning_rate": 1.2640000000000001e-05,
|
| 181 |
+
"loss": 1.4843,
|
| 182 |
+
"step": 230
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.92,
|
| 186 |
+
"grad_norm": 12.766644477844238,
|
| 187 |
+
"learning_rate": 1.232e-05,
|
| 188 |
+
"loss": 1.4018,
|
| 189 |
+
"step": 240
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.0,
|
| 193 |
+
"grad_norm": 7.949278831481934,
|
| 194 |
+
"learning_rate": 1.2e-05,
|
| 195 |
+
"loss": 1.5226,
|
| 196 |
+
"step": 250
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 2.0,
|
| 200 |
+
"eval_accuracy": 0.47391304347826085,
|
| 201 |
+
"eval_f1": 0.40774923443789374,
|
| 202 |
+
"eval_loss": 1.4421453475952148,
|
| 203 |
+
"eval_precision": 0.38951398452357533,
|
| 204 |
+
"eval_recall": 0.47391304347826085,
|
| 205 |
+
"eval_runtime": 1.2912,
|
| 206 |
+
"eval_samples_per_second": 178.127,
|
| 207 |
+
"eval_steps_per_second": 11.617,
|
| 208 |
+
"step": 250
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.08,
|
| 212 |
+
"grad_norm": 6.321002960205078,
|
| 213 |
+
"learning_rate": 1.168e-05,
|
| 214 |
+
"loss": 1.3278,
|
| 215 |
+
"step": 260
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 2.16,
|
| 219 |
+
"grad_norm": 5.3338093757629395,
|
| 220 |
+
"learning_rate": 1.136e-05,
|
| 221 |
+
"loss": 1.2814,
|
| 222 |
+
"step": 270
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.24,
|
| 226 |
+
"grad_norm": 4.96796178817749,
|
| 227 |
+
"learning_rate": 1.1040000000000001e-05,
|
| 228 |
+
"loss": 1.4562,
|
| 229 |
+
"step": 280
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 2.32,
|
| 233 |
+
"grad_norm": 10.834942817687988,
|
| 234 |
+
"learning_rate": 1.072e-05,
|
| 235 |
+
"loss": 1.39,
|
| 236 |
+
"step": 290
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.4,
|
| 240 |
+
"grad_norm": 8.158452033996582,
|
| 241 |
+
"learning_rate": 1.04e-05,
|
| 242 |
+
"loss": 1.3208,
|
| 243 |
+
"step": 300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.48,
|
| 247 |
+
"grad_norm": 6.426680088043213,
|
| 248 |
+
"learning_rate": 1.008e-05,
|
| 249 |
+
"loss": 1.3216,
|
| 250 |
+
"step": 310
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.56,
|
| 254 |
+
"grad_norm": 7.6382904052734375,
|
| 255 |
+
"learning_rate": 9.760000000000001e-06,
|
| 256 |
+
"loss": 1.3789,
|
| 257 |
+
"step": 320
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 2.64,
|
| 261 |
+
"grad_norm": 7.195258617401123,
|
| 262 |
+
"learning_rate": 9.440000000000001e-06,
|
| 263 |
+
"loss": 1.4057,
|
| 264 |
+
"step": 330
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 2.72,
|
| 268 |
+
"grad_norm": 6.953582763671875,
|
| 269 |
+
"learning_rate": 9.12e-06,
|
| 270 |
+
"loss": 1.328,
|
| 271 |
+
"step": 340
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 2.8,
|
| 275 |
+
"grad_norm": 7.441910743713379,
|
| 276 |
+
"learning_rate": 8.8e-06,
|
| 277 |
+
"loss": 1.3708,
|
| 278 |
+
"step": 350
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.88,
|
| 282 |
+
"grad_norm": 10.070989608764648,
|
| 283 |
+
"learning_rate": 8.48e-06,
|
| 284 |
+
"loss": 1.288,
|
| 285 |
+
"step": 360
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.96,
|
| 289 |
+
"grad_norm": 7.511960029602051,
|
| 290 |
+
"learning_rate": 8.16e-06,
|
| 291 |
+
"loss": 1.1656,
|
| 292 |
+
"step": 370
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 3.0,
|
| 296 |
+
"eval_accuracy": 0.5260869565217391,
|
| 297 |
+
"eval_f1": 0.479516001356244,
|
| 298 |
+
"eval_loss": 1.313248872756958,
|
| 299 |
+
"eval_precision": 0.448993952375781,
|
| 300 |
+
"eval_recall": 0.5260869565217391,
|
| 301 |
+
"eval_runtime": 1.2954,
|
| 302 |
+
"eval_samples_per_second": 177.546,
|
| 303 |
+
"eval_steps_per_second": 11.579,
|
| 304 |
+
"step": 375
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 3.04,
|
| 308 |
+
"grad_norm": 14.974254608154297,
|
| 309 |
+
"learning_rate": 7.840000000000001e-06,
|
| 310 |
+
"loss": 1.1554,
|
| 311 |
+
"step": 380
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 3.12,
|
| 315 |
+
"grad_norm": 6.2412238121032715,
|
| 316 |
+
"learning_rate": 7.520000000000001e-06,
|
| 317 |
+
"loss": 1.0172,
|
| 318 |
+
"step": 390
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 3.2,
|
| 322 |
+
"grad_norm": 8.691516876220703,
|
| 323 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 324 |
+
"loss": 1.3717,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 3.28,
|
| 329 |
+
"grad_norm": 8.422670364379883,
|
| 330 |
+
"learning_rate": 6.88e-06,
|
| 331 |
+
"loss": 1.187,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 3.36,
|
| 336 |
+
"grad_norm": 6.4102396965026855,
|
| 337 |
+
"learning_rate": 6.560000000000001e-06,
|
| 338 |
+
"loss": 1.0076,
|
| 339 |
+
"step": 420
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.44,
|
| 343 |
+
"grad_norm": 8.417737007141113,
|
| 344 |
+
"learning_rate": 6.24e-06,
|
| 345 |
+
"loss": 1.1928,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 3.52,
|
| 350 |
+
"grad_norm": 9.579270362854004,
|
| 351 |
+
"learning_rate": 5.92e-06,
|
| 352 |
+
"loss": 1.0687,
|
| 353 |
+
"step": 440
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 3.6,
|
| 357 |
+
"grad_norm": 11.200490951538086,
|
| 358 |
+
"learning_rate": 5.600000000000001e-06,
|
| 359 |
+
"loss": 1.11,
|
| 360 |
+
"step": 450
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 3.68,
|
| 364 |
+
"grad_norm": 9.393120765686035,
|
| 365 |
+
"learning_rate": 5.28e-06,
|
| 366 |
+
"loss": 1.1811,
|
| 367 |
+
"step": 460
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 3.76,
|
| 371 |
+
"grad_norm": 9.762161254882812,
|
| 372 |
+
"learning_rate": 4.960000000000001e-06,
|
| 373 |
+
"loss": 1.0488,
|
| 374 |
+
"step": 470
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 3.84,
|
| 378 |
+
"grad_norm": 10.627289772033691,
|
| 379 |
+
"learning_rate": 4.6400000000000005e-06,
|
| 380 |
+
"loss": 1.1537,
|
| 381 |
+
"step": 480
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.92,
|
| 385 |
+
"grad_norm": 13.637901306152344,
|
| 386 |
+
"learning_rate": 4.32e-06,
|
| 387 |
+
"loss": 1.032,
|
| 388 |
+
"step": 490
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 4.0,
|
| 392 |
+
"grad_norm": 13.00786018371582,
|
| 393 |
+
"learning_rate": 4.000000000000001e-06,
|
| 394 |
+
"loss": 1.1095,
|
| 395 |
+
"step": 500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 4.0,
|
| 399 |
+
"eval_accuracy": 0.5565217391304348,
|
| 400 |
+
"eval_f1": 0.5231441442614202,
|
| 401 |
+
"eval_loss": 1.2819163799285889,
|
| 402 |
+
"eval_precision": 0.5155867764196465,
|
| 403 |
+
"eval_recall": 0.5565217391304348,
|
| 404 |
+
"eval_runtime": 1.2834,
|
| 405 |
+
"eval_samples_per_second": 179.211,
|
| 406 |
+
"eval_steps_per_second": 11.688,
|
| 407 |
+
"step": 500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 4.08,
|
| 411 |
+
"grad_norm": 9.446922302246094,
|
| 412 |
+
"learning_rate": 3.6800000000000003e-06,
|
| 413 |
+
"loss": 0.88,
|
| 414 |
+
"step": 510
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.16,
|
| 418 |
+
"grad_norm": 9.645947456359863,
|
| 419 |
+
"learning_rate": 3.3600000000000004e-06,
|
| 420 |
+
"loss": 0.9479,
|
| 421 |
+
"step": 520
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 4.24,
|
| 425 |
+
"grad_norm": 6.6881184577941895,
|
| 426 |
+
"learning_rate": 3.04e-06,
|
| 427 |
+
"loss": 1.0034,
|
| 428 |
+
"step": 530
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.32,
|
| 432 |
+
"grad_norm": 13.422988891601562,
|
| 433 |
+
"learning_rate": 2.7200000000000002e-06,
|
| 434 |
+
"loss": 1.0929,
|
| 435 |
+
"step": 540
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.4,
|
| 439 |
+
"grad_norm": 9.193038940429688,
|
| 440 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 441 |
+
"loss": 0.9017,
|
| 442 |
+
"step": 550
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 4.48,
|
| 446 |
+
"grad_norm": 8.080782890319824,
|
| 447 |
+
"learning_rate": 2.08e-06,
|
| 448 |
+
"loss": 0.9509,
|
| 449 |
+
"step": 560
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 4.56,
|
| 453 |
+
"grad_norm": 10.08934497833252,
|
| 454 |
+
"learning_rate": 1.76e-06,
|
| 455 |
+
"loss": 0.9522,
|
| 456 |
+
"step": 570
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 4.64,
|
| 460 |
+
"grad_norm": 8.592775344848633,
|
| 461 |
+
"learning_rate": 1.44e-06,
|
| 462 |
+
"loss": 0.9903,
|
| 463 |
+
"step": 580
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 4.72,
|
| 467 |
+
"grad_norm": 10.12938117980957,
|
| 468 |
+
"learning_rate": 1.12e-06,
|
| 469 |
+
"loss": 0.9153,
|
| 470 |
+
"step": 590
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.8,
|
| 474 |
+
"grad_norm": 13.647997856140137,
|
| 475 |
+
"learning_rate": 8.000000000000001e-07,
|
| 476 |
+
"loss": 1.0831,
|
| 477 |
+
"step": 600
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 4.88,
|
| 481 |
+
"grad_norm": 15.894301414489746,
|
| 482 |
+
"learning_rate": 4.800000000000001e-07,
|
| 483 |
+
"loss": 1.1632,
|
| 484 |
+
"step": 610
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 4.96,
|
| 488 |
+
"grad_norm": 11.063687324523926,
|
| 489 |
+
"learning_rate": 1.6e-07,
|
| 490 |
+
"loss": 1.0974,
|
| 491 |
+
"step": 620
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 5.0,
|
| 495 |
+
"eval_accuracy": 0.5565217391304348,
|
| 496 |
+
"eval_f1": 0.5303004793794206,
|
| 497 |
+
"eval_loss": 1.269640326499939,
|
| 498 |
+
"eval_precision": 0.532996572682937,
|
| 499 |
+
"eval_recall": 0.5565217391304348,
|
| 500 |
+
"eval_runtime": 1.2978,
|
| 501 |
+
"eval_samples_per_second": 177.219,
|
| 502 |
+
"eval_steps_per_second": 11.558,
|
| 503 |
+
"step": 625
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 5.0,
|
| 507 |
+
"step": 625,
|
| 508 |
+
"total_flos": 1434186246250944.0,
|
| 509 |
+
"train_loss": 1.3634769760131835,
|
| 510 |
+
"train_runtime": 225.3467,
|
| 511 |
+
"train_samples_per_second": 44.287,
|
| 512 |
+
"train_steps_per_second": 2.774
|
| 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
|
| 524 |
+
}
|