End of training
Browse files- README.md +117 -0
- all_results.json +12 -0
- config.json +79 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +28 -0
- runs/Dec04_08-31-58_DESKTOP-SKBE9FB/events.out.tfevents.1733322720.DESKTOP-SKBE9FB.15988.0 +3 -0
- runs/Dec04_08-31-58_DESKTOP-SKBE9FB/events.out.tfevents.1733326971.DESKTOP-SKBE9FB.15988.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +2694 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: microsoft/beit-base-patch16-224
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- imagefolder
|
| 8 |
+
metrics:
|
| 9 |
+
- accuracy
|
| 10 |
+
model-index:
|
| 11 |
+
- name: BEiT-RD-DA
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Image Classification
|
| 15 |
+
type: image-classification
|
| 16 |
+
dataset:
|
| 17 |
+
name: imagefolder
|
| 18 |
+
type: imagefolder
|
| 19 |
+
config: default
|
| 20 |
+
split: validation
|
| 21 |
+
args: default
|
| 22 |
+
metrics:
|
| 23 |
+
- name: Accuracy
|
| 24 |
+
type: accuracy
|
| 25 |
+
value: 0.6654545454545454
|
| 26 |
+
---
|
| 27 |
+
|
| 28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 29 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 30 |
+
|
| 31 |
+
# BEiT-RD-DA
|
| 32 |
+
|
| 33 |
+
This model is a fine-tuned version of [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
+
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 1.9617
|
| 36 |
+
- Accuracy: 0.6655
|
| 37 |
+
|
| 38 |
+
## Model description
|
| 39 |
+
|
| 40 |
+
More information needed
|
| 41 |
+
|
| 42 |
+
## Intended uses & limitations
|
| 43 |
+
|
| 44 |
+
More information needed
|
| 45 |
+
|
| 46 |
+
## Training and evaluation data
|
| 47 |
+
|
| 48 |
+
More information needed
|
| 49 |
+
|
| 50 |
+
## Training procedure
|
| 51 |
+
|
| 52 |
+
### Training hyperparameters
|
| 53 |
+
|
| 54 |
+
The following hyperparameters were used during training:
|
| 55 |
+
- learning_rate: 5e-05
|
| 56 |
+
- train_batch_size: 16
|
| 57 |
+
- eval_batch_size: 16
|
| 58 |
+
- seed: 42
|
| 59 |
+
- gradient_accumulation_steps: 4
|
| 60 |
+
- total_train_batch_size: 64
|
| 61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
+
- lr_scheduler_type: linear
|
| 63 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 64 |
+
- num_epochs: 40
|
| 65 |
+
|
| 66 |
+
### Training results
|
| 67 |
+
|
| 68 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 69 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 70 |
+
| 1.4123 | 1.0 | 96 | 1.4099 | 0.4927 |
|
| 71 |
+
| 0.9503 | 2.0 | 192 | 1.8852 | 0.4927 |
|
| 72 |
+
| 0.8284 | 3.0 | 288 | 2.1702 | 0.5073 |
|
| 73 |
+
| 0.7677 | 4.0 | 384 | 2.0408 | 0.5345 |
|
| 74 |
+
| 0.788 | 5.0 | 480 | 2.7991 | 0.5127 |
|
| 75 |
+
| 0.5822 | 6.0 | 576 | 2.0951 | 0.5636 |
|
| 76 |
+
| 0.5172 | 7.0 | 672 | 2.5977 | 0.5364 |
|
| 77 |
+
| 0.4615 | 8.0 | 768 | 2.0968 | 0.58 |
|
| 78 |
+
| 0.3672 | 9.0 | 864 | 2.8535 | 0.5436 |
|
| 79 |
+
| 0.379 | 10.0 | 960 | 2.9515 | 0.5382 |
|
| 80 |
+
| 0.3301 | 11.0 | 1056 | 2.7200 | 0.5582 |
|
| 81 |
+
| 0.2786 | 12.0 | 1152 | 1.9000 | 0.6273 |
|
| 82 |
+
| 0.2746 | 13.0 | 1248 | 3.1768 | 0.5364 |
|
| 83 |
+
| 0.2298 | 14.0 | 1344 | 3.1003 | 0.5527 |
|
| 84 |
+
| 0.2013 | 15.0 | 1440 | 2.3441 | 0.6182 |
|
| 85 |
+
| 0.2225 | 16.0 | 1536 | 3.0214 | 0.5709 |
|
| 86 |
+
| 0.2229 | 17.0 | 1632 | 2.0676 | 0.6164 |
|
| 87 |
+
| 0.2024 | 18.0 | 1728 | 2.6478 | 0.5673 |
|
| 88 |
+
| 0.1401 | 19.0 | 1824 | 2.8952 | 0.5636 |
|
| 89 |
+
| 0.1984 | 20.0 | 1920 | 2.3083 | 0.6145 |
|
| 90 |
+
| 0.1788 | 21.0 | 2016 | 3.7702 | 0.52 |
|
| 91 |
+
| 0.1907 | 22.0 | 2112 | 1.9617 | 0.6655 |
|
| 92 |
+
| 0.1113 | 23.0 | 2208 | 2.6546 | 0.5964 |
|
| 93 |
+
| 0.1293 | 24.0 | 2304 | 2.6427 | 0.6036 |
|
| 94 |
+
| 0.1354 | 25.0 | 2400 | 3.4105 | 0.5527 |
|
| 95 |
+
| 0.1447 | 26.0 | 2496 | 2.5460 | 0.6127 |
|
| 96 |
+
| 0.0995 | 27.0 | 2592 | 2.9865 | 0.5855 |
|
| 97 |
+
| 0.1369 | 28.0 | 2688 | 3.5281 | 0.5545 |
|
| 98 |
+
| 0.1238 | 29.0 | 2784 | 2.8161 | 0.6018 |
|
| 99 |
+
| 0.1256 | 30.0 | 2880 | 3.4917 | 0.5491 |
|
| 100 |
+
| 0.1064 | 31.0 | 2976 | 3.0659 | 0.58 |
|
| 101 |
+
| 0.1333 | 32.0 | 3072 | 3.5972 | 0.5473 |
|
| 102 |
+
| 0.1134 | 33.0 | 3168 | 3.6116 | 0.54 |
|
| 103 |
+
| 0.0831 | 34.0 | 3264 | 3.5308 | 0.5509 |
|
| 104 |
+
| 0.1035 | 35.0 | 3360 | 3.4789 | 0.5582 |
|
| 105 |
+
| 0.0957 | 36.0 | 3456 | 3.6358 | 0.5509 |
|
| 106 |
+
| 0.0764 | 37.0 | 3552 | 3.3639 | 0.5709 |
|
| 107 |
+
| 0.072 | 38.0 | 3648 | 3.5639 | 0.5564 |
|
| 108 |
+
| 0.0727 | 39.0 | 3744 | 3.5193 | 0.5582 |
|
| 109 |
+
| 0.0619 | 40.0 | 3840 | 3.5836 | 0.5582 |
|
| 110 |
+
|
| 111 |
+
|
| 112 |
+
### Framework versions
|
| 113 |
+
|
| 114 |
+
- Transformers 4.36.2
|
| 115 |
+
- Pytorch 2.1.2+cu118
|
| 116 |
+
- Datasets 2.16.1
|
| 117 |
+
- Tokenizers 0.15.0
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 0.6654545454545454,
|
| 4 |
+
"eval_loss": 1.9616814851760864,
|
| 5 |
+
"eval_runtime": 4.2986,
|
| 6 |
+
"eval_samples_per_second": 127.949,
|
| 7 |
+
"eval_steps_per_second": 8.142,
|
| 8 |
+
"train_loss": 0.31003439707371094,
|
| 9 |
+
"train_runtime": 4246.6951,
|
| 10 |
+
"train_samples_per_second": 57.767,
|
| 11 |
+
"train_steps_per_second": 0.904
|
| 12 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/beit-base-patch16-224",
|
| 3 |
+
"add_fpn": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"BeitForImageClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_probs_dropout_prob": 0.0,
|
| 8 |
+
"auxiliary_channels": 256,
|
| 9 |
+
"auxiliary_concat_input": false,
|
| 10 |
+
"auxiliary_loss_weight": 0.4,
|
| 11 |
+
"auxiliary_num_convs": 1,
|
| 12 |
+
"drop_path_rate": 0.1,
|
| 13 |
+
"hidden_act": "gelu",
|
| 14 |
+
"hidden_dropout_prob": 0.0,
|
| 15 |
+
"hidden_size": 768,
|
| 16 |
+
"id2label": {
|
| 17 |
+
"0": "Mild",
|
| 18 |
+
"1": "Moderate",
|
| 19 |
+
"2": "No_DR",
|
| 20 |
+
"3": "Proliferate_DR",
|
| 21 |
+
"4": "Severe"
|
| 22 |
+
},
|
| 23 |
+
"image_size": 224,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"intermediate_size": 3072,
|
| 26 |
+
"label2id": {
|
| 27 |
+
"Mild": 0,
|
| 28 |
+
"Moderate": 1,
|
| 29 |
+
"No_DR": 2,
|
| 30 |
+
"Proliferate_DR": 3,
|
| 31 |
+
"Severe": 4
|
| 32 |
+
},
|
| 33 |
+
"layer_norm_eps": 1e-12,
|
| 34 |
+
"layer_scale_init_value": 0.1,
|
| 35 |
+
"model_type": "beit",
|
| 36 |
+
"num_attention_heads": 12,
|
| 37 |
+
"num_channels": 3,
|
| 38 |
+
"num_hidden_layers": 12,
|
| 39 |
+
"out_features": [
|
| 40 |
+
"stage12"
|
| 41 |
+
],
|
| 42 |
+
"out_indices": [
|
| 43 |
+
12
|
| 44 |
+
],
|
| 45 |
+
"patch_size": 16,
|
| 46 |
+
"pool_scales": [
|
| 47 |
+
1,
|
| 48 |
+
2,
|
| 49 |
+
3,
|
| 50 |
+
6
|
| 51 |
+
],
|
| 52 |
+
"problem_type": "single_label_classification",
|
| 53 |
+
"reshape_hidden_states": true,
|
| 54 |
+
"semantic_loss_ignore_index": 255,
|
| 55 |
+
"stage_names": [
|
| 56 |
+
"stem",
|
| 57 |
+
"stage1",
|
| 58 |
+
"stage2",
|
| 59 |
+
"stage3",
|
| 60 |
+
"stage4",
|
| 61 |
+
"stage5",
|
| 62 |
+
"stage6",
|
| 63 |
+
"stage7",
|
| 64 |
+
"stage8",
|
| 65 |
+
"stage9",
|
| 66 |
+
"stage10",
|
| 67 |
+
"stage11",
|
| 68 |
+
"stage12"
|
| 69 |
+
],
|
| 70 |
+
"torch_dtype": "float32",
|
| 71 |
+
"transformers_version": "4.36.2",
|
| 72 |
+
"use_absolute_position_embeddings": false,
|
| 73 |
+
"use_auxiliary_head": true,
|
| 74 |
+
"use_mask_token": false,
|
| 75 |
+
"use_mean_pooling": true,
|
| 76 |
+
"use_relative_position_bias": true,
|
| 77 |
+
"use_shared_relative_position_bias": false,
|
| 78 |
+
"vocab_size": 8192
|
| 79 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 0.6654545454545454,
|
| 4 |
+
"eval_loss": 1.9616814851760864,
|
| 5 |
+
"eval_runtime": 4.2986,
|
| 6 |
+
"eval_samples_per_second": 127.949,
|
| 7 |
+
"eval_steps_per_second": 8.142
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3809e4e83e75214f6476efc19124c3e2d6df7a18a8edb66a818365824341372
|
| 3 |
+
size 343089556
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": {
|
| 3 |
+
"height": 224,
|
| 4 |
+
"width": 224
|
| 5 |
+
},
|
| 6 |
+
"do_center_crop": false,
|
| 7 |
+
"do_normalize": true,
|
| 8 |
+
"do_reduce_labels": false,
|
| 9 |
+
"do_rescale": true,
|
| 10 |
+
"do_resize": true,
|
| 11 |
+
"image_mean": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"image_processor_type": "BeitImageProcessor",
|
| 17 |
+
"image_std": [
|
| 18 |
+
0.5,
|
| 19 |
+
0.5,
|
| 20 |
+
0.5
|
| 21 |
+
],
|
| 22 |
+
"resample": 2,
|
| 23 |
+
"rescale_factor": 0.00392156862745098,
|
| 24 |
+
"size": {
|
| 25 |
+
"height": 224,
|
| 26 |
+
"width": 224
|
| 27 |
+
}
|
| 28 |
+
}
|
runs/Dec04_08-31-58_DESKTOP-SKBE9FB/events.out.tfevents.1733322720.DESKTOP-SKBE9FB.15988.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a4de50255b52e76f890337495e63561df538068bf66506a5d5ba0e81dc938257
|
| 3 |
+
size 78825
|
runs/Dec04_08-31-58_DESKTOP-SKBE9FB/events.out.tfevents.1733326971.DESKTOP-SKBE9FB.15988.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de120b81bade123be7bb882d92c13a37823561c353d8f8d182d0d3dd51d0c778
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"train_loss": 0.31003439707371094,
|
| 4 |
+
"train_runtime": 4246.6951,
|
| 5 |
+
"train_samples_per_second": 57.767,
|
| 6 |
+
"train_steps_per_second": 0.904
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2694 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6654545454545454,
|
| 3 |
+
"best_model_checkpoint": "BEiT-RD-DA\\checkpoint-2112",
|
| 4 |
+
"epoch": 40.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 3840,
|
| 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": 1.3020833333333335e-06,
|
| 14 |
+
"loss": 1.6093,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.21,
|
| 19 |
+
"learning_rate": 2.604166666666667e-06,
|
| 20 |
+
"loss": 1.6082,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.31,
|
| 25 |
+
"learning_rate": 3.90625e-06,
|
| 26 |
+
"loss": 1.6045,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.42,
|
| 31 |
+
"learning_rate": 5.208333333333334e-06,
|
| 32 |
+
"loss": 1.5939,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.52,
|
| 37 |
+
"learning_rate": 6.510416666666667e-06,
|
| 38 |
+
"loss": 1.576,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.62,
|
| 43 |
+
"learning_rate": 7.8125e-06,
|
| 44 |
+
"loss": 1.5501,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.73,
|
| 49 |
+
"learning_rate": 9.114583333333334e-06,
|
| 50 |
+
"loss": 1.5116,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.83,
|
| 55 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 56 |
+
"loss": 1.462,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.94,
|
| 61 |
+
"learning_rate": 1.171875e-05,
|
| 62 |
+
"loss": 1.4123,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.0,
|
| 67 |
+
"eval_accuracy": 0.49272727272727274,
|
| 68 |
+
"eval_loss": 1.4099408388137817,
|
| 69 |
+
"eval_runtime": 7.0436,
|
| 70 |
+
"eval_samples_per_second": 78.085,
|
| 71 |
+
"eval_steps_per_second": 4.969,
|
| 72 |
+
"step": 96
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.04,
|
| 76 |
+
"learning_rate": 1.3020833333333334e-05,
|
| 77 |
+
"loss": 1.3572,
|
| 78 |
+
"step": 100
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.15,
|
| 82 |
+
"learning_rate": 1.4322916666666666e-05,
|
| 83 |
+
"loss": 1.2997,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.25,
|
| 88 |
+
"learning_rate": 1.5625e-05,
|
| 89 |
+
"loss": 1.2359,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.35,
|
| 94 |
+
"learning_rate": 1.6927083333333336e-05,
|
| 95 |
+
"loss": 1.1928,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.46,
|
| 100 |
+
"learning_rate": 1.8229166666666668e-05,
|
| 101 |
+
"loss": 1.1188,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.56,
|
| 106 |
+
"learning_rate": 1.953125e-05,
|
| 107 |
+
"loss": 1.0731,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.67,
|
| 112 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 113 |
+
"loss": 1.0396,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.77,
|
| 118 |
+
"learning_rate": 2.2135416666666668e-05,
|
| 119 |
+
"loss": 0.9933,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.88,
|
| 124 |
+
"learning_rate": 2.34375e-05,
|
| 125 |
+
"loss": 0.9498,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.98,
|
| 130 |
+
"learning_rate": 2.4739583333333336e-05,
|
| 131 |
+
"loss": 0.9503,
|
| 132 |
+
"step": 190
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.0,
|
| 136 |
+
"eval_accuracy": 0.49272727272727274,
|
| 137 |
+
"eval_loss": 1.8851675987243652,
|
| 138 |
+
"eval_runtime": 3.6814,
|
| 139 |
+
"eval_samples_per_second": 149.401,
|
| 140 |
+
"eval_steps_per_second": 9.507,
|
| 141 |
+
"step": 192
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 2.08,
|
| 145 |
+
"learning_rate": 2.604166666666667e-05,
|
| 146 |
+
"loss": 0.9379,
|
| 147 |
+
"step": 200
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.19,
|
| 151 |
+
"learning_rate": 2.734375e-05,
|
| 152 |
+
"loss": 0.8963,
|
| 153 |
+
"step": 210
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.29,
|
| 157 |
+
"learning_rate": 2.8645833333333333e-05,
|
| 158 |
+
"loss": 0.9422,
|
| 159 |
+
"step": 220
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.4,
|
| 163 |
+
"learning_rate": 2.994791666666667e-05,
|
| 164 |
+
"loss": 0.904,
|
| 165 |
+
"step": 230
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.5,
|
| 169 |
+
"learning_rate": 3.125e-05,
|
| 170 |
+
"loss": 0.9131,
|
| 171 |
+
"step": 240
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.6,
|
| 175 |
+
"learning_rate": 3.255208333333333e-05,
|
| 176 |
+
"loss": 0.9078,
|
| 177 |
+
"step": 250
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.71,
|
| 181 |
+
"learning_rate": 3.385416666666667e-05,
|
| 182 |
+
"loss": 0.851,
|
| 183 |
+
"step": 260
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 2.81,
|
| 187 |
+
"learning_rate": 3.5156250000000004e-05,
|
| 188 |
+
"loss": 0.9054,
|
| 189 |
+
"step": 270
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.92,
|
| 193 |
+
"learning_rate": 3.6458333333333336e-05,
|
| 194 |
+
"loss": 0.8284,
|
| 195 |
+
"step": 280
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.0,
|
| 199 |
+
"eval_accuracy": 0.5072727272727273,
|
| 200 |
+
"eval_loss": 2.1701955795288086,
|
| 201 |
+
"eval_runtime": 3.6684,
|
| 202 |
+
"eval_samples_per_second": 149.93,
|
| 203 |
+
"eval_steps_per_second": 9.541,
|
| 204 |
+
"step": 288
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.02,
|
| 208 |
+
"learning_rate": 3.776041666666667e-05,
|
| 209 |
+
"loss": 0.7817,
|
| 210 |
+
"step": 290
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 3.12,
|
| 214 |
+
"learning_rate": 3.90625e-05,
|
| 215 |
+
"loss": 0.8264,
|
| 216 |
+
"step": 300
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 3.23,
|
| 220 |
+
"learning_rate": 4.036458333333333e-05,
|
| 221 |
+
"loss": 0.8271,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.33,
|
| 226 |
+
"learning_rate": 4.166666666666667e-05,
|
| 227 |
+
"loss": 0.8185,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.44,
|
| 232 |
+
"learning_rate": 4.2968750000000004e-05,
|
| 233 |
+
"loss": 0.7828,
|
| 234 |
+
"step": 330
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 3.54,
|
| 238 |
+
"learning_rate": 4.4270833333333337e-05,
|
| 239 |
+
"loss": 0.7351,
|
| 240 |
+
"step": 340
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.65,
|
| 244 |
+
"learning_rate": 4.557291666666667e-05,
|
| 245 |
+
"loss": 0.7248,
|
| 246 |
+
"step": 350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.75,
|
| 250 |
+
"learning_rate": 4.6875e-05,
|
| 251 |
+
"loss": 0.7981,
|
| 252 |
+
"step": 360
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 3.85,
|
| 256 |
+
"learning_rate": 4.817708333333333e-05,
|
| 257 |
+
"loss": 0.8329,
|
| 258 |
+
"step": 370
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 3.96,
|
| 262 |
+
"learning_rate": 4.947916666666667e-05,
|
| 263 |
+
"loss": 0.7677,
|
| 264 |
+
"step": 380
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 4.0,
|
| 268 |
+
"eval_accuracy": 0.5345454545454545,
|
| 269 |
+
"eval_loss": 2.0407912731170654,
|
| 270 |
+
"eval_runtime": 4.124,
|
| 271 |
+
"eval_samples_per_second": 133.366,
|
| 272 |
+
"eval_steps_per_second": 8.487,
|
| 273 |
+
"step": 384
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 4.06,
|
| 277 |
+
"learning_rate": 4.991319444444445e-05,
|
| 278 |
+
"loss": 0.7784,
|
| 279 |
+
"step": 390
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 4.17,
|
| 283 |
+
"learning_rate": 4.976851851851852e-05,
|
| 284 |
+
"loss": 0.7742,
|
| 285 |
+
"step": 400
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.27,
|
| 289 |
+
"learning_rate": 4.96238425925926e-05,
|
| 290 |
+
"loss": 0.7786,
|
| 291 |
+
"step": 410
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 4.38,
|
| 295 |
+
"learning_rate": 4.947916666666667e-05,
|
| 296 |
+
"loss": 0.7264,
|
| 297 |
+
"step": 420
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.48,
|
| 301 |
+
"learning_rate": 4.933449074074074e-05,
|
| 302 |
+
"loss": 0.752,
|
| 303 |
+
"step": 430
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 4.58,
|
| 307 |
+
"learning_rate": 4.9189814814814815e-05,
|
| 308 |
+
"loss": 0.7815,
|
| 309 |
+
"step": 440
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 4.69,
|
| 313 |
+
"learning_rate": 4.904513888888889e-05,
|
| 314 |
+
"loss": 0.7081,
|
| 315 |
+
"step": 450
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 4.79,
|
| 319 |
+
"learning_rate": 4.8900462962962965e-05,
|
| 320 |
+
"loss": 0.7476,
|
| 321 |
+
"step": 460
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 4.9,
|
| 325 |
+
"learning_rate": 4.875578703703704e-05,
|
| 326 |
+
"loss": 0.6773,
|
| 327 |
+
"step": 470
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 5.0,
|
| 331 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 332 |
+
"loss": 0.788,
|
| 333 |
+
"step": 480
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 5.0,
|
| 337 |
+
"eval_accuracy": 0.5127272727272727,
|
| 338 |
+
"eval_loss": 2.7991154193878174,
|
| 339 |
+
"eval_runtime": 3.8459,
|
| 340 |
+
"eval_samples_per_second": 143.009,
|
| 341 |
+
"eval_steps_per_second": 9.101,
|
| 342 |
+
"step": 480
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 5.1,
|
| 346 |
+
"learning_rate": 4.846643518518519e-05,
|
| 347 |
+
"loss": 0.7058,
|
| 348 |
+
"step": 490
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 5.21,
|
| 352 |
+
"learning_rate": 4.8321759259259265e-05,
|
| 353 |
+
"loss": 0.7083,
|
| 354 |
+
"step": 500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 5.31,
|
| 358 |
+
"learning_rate": 4.817708333333333e-05,
|
| 359 |
+
"loss": 0.6479,
|
| 360 |
+
"step": 510
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 5.42,
|
| 364 |
+
"learning_rate": 4.803240740740741e-05,
|
| 365 |
+
"loss": 0.658,
|
| 366 |
+
"step": 520
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 5.52,
|
| 370 |
+
"learning_rate": 4.788773148148148e-05,
|
| 371 |
+
"loss": 0.7756,
|
| 372 |
+
"step": 530
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 5.62,
|
| 376 |
+
"learning_rate": 4.774305555555556e-05,
|
| 377 |
+
"loss": 0.6294,
|
| 378 |
+
"step": 540
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 5.73,
|
| 382 |
+
"learning_rate": 4.759837962962963e-05,
|
| 383 |
+
"loss": 0.6727,
|
| 384 |
+
"step": 550
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 5.83,
|
| 388 |
+
"learning_rate": 4.745370370370371e-05,
|
| 389 |
+
"loss": 0.6476,
|
| 390 |
+
"step": 560
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 5.94,
|
| 394 |
+
"learning_rate": 4.730902777777778e-05,
|
| 395 |
+
"loss": 0.5822,
|
| 396 |
+
"step": 570
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 6.0,
|
| 400 |
+
"eval_accuracy": 0.5636363636363636,
|
| 401 |
+
"eval_loss": 2.0950703620910645,
|
| 402 |
+
"eval_runtime": 3.9745,
|
| 403 |
+
"eval_samples_per_second": 138.384,
|
| 404 |
+
"eval_steps_per_second": 8.806,
|
| 405 |
+
"step": 576
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 6.04,
|
| 409 |
+
"learning_rate": 4.716435185185186e-05,
|
| 410 |
+
"loss": 0.5656,
|
| 411 |
+
"step": 580
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 6.15,
|
| 415 |
+
"learning_rate": 4.7019675925925926e-05,
|
| 416 |
+
"loss": 0.6338,
|
| 417 |
+
"step": 590
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 6.25,
|
| 421 |
+
"learning_rate": 4.6875e-05,
|
| 422 |
+
"loss": 0.62,
|
| 423 |
+
"step": 600
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 6.35,
|
| 427 |
+
"learning_rate": 4.6730324074074076e-05,
|
| 428 |
+
"loss": 0.5603,
|
| 429 |
+
"step": 610
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 6.46,
|
| 433 |
+
"learning_rate": 4.658564814814815e-05,
|
| 434 |
+
"loss": 0.5726,
|
| 435 |
+
"step": 620
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 6.56,
|
| 439 |
+
"learning_rate": 4.6440972222222226e-05,
|
| 440 |
+
"loss": 0.5767,
|
| 441 |
+
"step": 630
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.67,
|
| 445 |
+
"learning_rate": 4.62962962962963e-05,
|
| 446 |
+
"loss": 0.5546,
|
| 447 |
+
"step": 640
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 6.77,
|
| 451 |
+
"learning_rate": 4.6151620370370376e-05,
|
| 452 |
+
"loss": 0.5003,
|
| 453 |
+
"step": 650
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 6.88,
|
| 457 |
+
"learning_rate": 4.6006944444444444e-05,
|
| 458 |
+
"loss": 0.564,
|
| 459 |
+
"step": 660
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 6.98,
|
| 463 |
+
"learning_rate": 4.586226851851852e-05,
|
| 464 |
+
"loss": 0.5172,
|
| 465 |
+
"step": 670
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 7.0,
|
| 469 |
+
"eval_accuracy": 0.5363636363636364,
|
| 470 |
+
"eval_loss": 2.5976808071136475,
|
| 471 |
+
"eval_runtime": 4.1525,
|
| 472 |
+
"eval_samples_per_second": 132.451,
|
| 473 |
+
"eval_steps_per_second": 8.429,
|
| 474 |
+
"step": 672
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 7.08,
|
| 478 |
+
"learning_rate": 4.5717592592592594e-05,
|
| 479 |
+
"loss": 0.4963,
|
| 480 |
+
"step": 680
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 7.19,
|
| 484 |
+
"learning_rate": 4.557291666666667e-05,
|
| 485 |
+
"loss": 0.5223,
|
| 486 |
+
"step": 690
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 7.29,
|
| 490 |
+
"learning_rate": 4.5428240740740744e-05,
|
| 491 |
+
"loss": 0.5279,
|
| 492 |
+
"step": 700
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 7.4,
|
| 496 |
+
"learning_rate": 4.528356481481482e-05,
|
| 497 |
+
"loss": 0.4685,
|
| 498 |
+
"step": 710
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 7.5,
|
| 502 |
+
"learning_rate": 4.5138888888888894e-05,
|
| 503 |
+
"loss": 0.4652,
|
| 504 |
+
"step": 720
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 7.6,
|
| 508 |
+
"learning_rate": 4.499421296296297e-05,
|
| 509 |
+
"loss": 0.3945,
|
| 510 |
+
"step": 730
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 7.71,
|
| 514 |
+
"learning_rate": 4.484953703703704e-05,
|
| 515 |
+
"loss": 0.4617,
|
| 516 |
+
"step": 740
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 7.81,
|
| 520 |
+
"learning_rate": 4.470486111111111e-05,
|
| 521 |
+
"loss": 0.471,
|
| 522 |
+
"step": 750
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 7.92,
|
| 526 |
+
"learning_rate": 4.456018518518519e-05,
|
| 527 |
+
"loss": 0.4615,
|
| 528 |
+
"step": 760
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 8.0,
|
| 532 |
+
"eval_accuracy": 0.58,
|
| 533 |
+
"eval_loss": 2.096754789352417,
|
| 534 |
+
"eval_runtime": 4.078,
|
| 535 |
+
"eval_samples_per_second": 134.871,
|
| 536 |
+
"eval_steps_per_second": 8.583,
|
| 537 |
+
"step": 768
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 8.02,
|
| 541 |
+
"learning_rate": 4.441550925925926e-05,
|
| 542 |
+
"loss": 0.4487,
|
| 543 |
+
"step": 770
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 8.12,
|
| 547 |
+
"learning_rate": 4.4270833333333337e-05,
|
| 548 |
+
"loss": 0.4692,
|
| 549 |
+
"step": 780
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 8.23,
|
| 553 |
+
"learning_rate": 4.412615740740741e-05,
|
| 554 |
+
"loss": 0.4161,
|
| 555 |
+
"step": 790
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 8.33,
|
| 559 |
+
"learning_rate": 4.3981481481481486e-05,
|
| 560 |
+
"loss": 0.4709,
|
| 561 |
+
"step": 800
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 8.44,
|
| 565 |
+
"learning_rate": 4.383680555555556e-05,
|
| 566 |
+
"loss": 0.4199,
|
| 567 |
+
"step": 810
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 8.54,
|
| 571 |
+
"learning_rate": 4.369212962962963e-05,
|
| 572 |
+
"loss": 0.4416,
|
| 573 |
+
"step": 820
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 8.65,
|
| 577 |
+
"learning_rate": 4.3547453703703704e-05,
|
| 578 |
+
"loss": 0.4253,
|
| 579 |
+
"step": 830
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 8.75,
|
| 583 |
+
"learning_rate": 4.340277777777778e-05,
|
| 584 |
+
"loss": 0.3953,
|
| 585 |
+
"step": 840
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 8.85,
|
| 589 |
+
"learning_rate": 4.3258101851851854e-05,
|
| 590 |
+
"loss": 0.4495,
|
| 591 |
+
"step": 850
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 8.96,
|
| 595 |
+
"learning_rate": 4.311342592592593e-05,
|
| 596 |
+
"loss": 0.3672,
|
| 597 |
+
"step": 860
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 9.0,
|
| 601 |
+
"eval_accuracy": 0.5436363636363636,
|
| 602 |
+
"eval_loss": 2.8535282611846924,
|
| 603 |
+
"eval_runtime": 3.6989,
|
| 604 |
+
"eval_samples_per_second": 148.694,
|
| 605 |
+
"eval_steps_per_second": 9.462,
|
| 606 |
+
"step": 864
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 9.06,
|
| 610 |
+
"learning_rate": 4.2968750000000004e-05,
|
| 611 |
+
"loss": 0.3812,
|
| 612 |
+
"step": 870
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 9.17,
|
| 616 |
+
"learning_rate": 4.282407407407408e-05,
|
| 617 |
+
"loss": 0.3683,
|
| 618 |
+
"step": 880
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 9.27,
|
| 622 |
+
"learning_rate": 4.267939814814815e-05,
|
| 623 |
+
"loss": 0.4462,
|
| 624 |
+
"step": 890
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 9.38,
|
| 628 |
+
"learning_rate": 4.253472222222222e-05,
|
| 629 |
+
"loss": 0.31,
|
| 630 |
+
"step": 900
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 9.48,
|
| 634 |
+
"learning_rate": 4.23900462962963e-05,
|
| 635 |
+
"loss": 0.3963,
|
| 636 |
+
"step": 910
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 9.58,
|
| 640 |
+
"learning_rate": 4.224537037037037e-05,
|
| 641 |
+
"loss": 0.3393,
|
| 642 |
+
"step": 920
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 9.69,
|
| 646 |
+
"learning_rate": 4.210069444444445e-05,
|
| 647 |
+
"loss": 0.3853,
|
| 648 |
+
"step": 930
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 9.79,
|
| 652 |
+
"learning_rate": 4.195601851851852e-05,
|
| 653 |
+
"loss": 0.4014,
|
| 654 |
+
"step": 940
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 9.9,
|
| 658 |
+
"learning_rate": 4.18113425925926e-05,
|
| 659 |
+
"loss": 0.3931,
|
| 660 |
+
"step": 950
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 10.0,
|
| 664 |
+
"learning_rate": 4.166666666666667e-05,
|
| 665 |
+
"loss": 0.379,
|
| 666 |
+
"step": 960
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 10.0,
|
| 670 |
+
"eval_accuracy": 0.5381818181818182,
|
| 671 |
+
"eval_loss": 2.9514999389648438,
|
| 672 |
+
"eval_runtime": 3.6769,
|
| 673 |
+
"eval_samples_per_second": 149.583,
|
| 674 |
+
"eval_steps_per_second": 9.519,
|
| 675 |
+
"step": 960
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 10.1,
|
| 679 |
+
"learning_rate": 4.152199074074074e-05,
|
| 680 |
+
"loss": 0.3544,
|
| 681 |
+
"step": 970
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 10.21,
|
| 685 |
+
"learning_rate": 4.1377314814814815e-05,
|
| 686 |
+
"loss": 0.3334,
|
| 687 |
+
"step": 980
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 10.31,
|
| 691 |
+
"learning_rate": 4.123263888888889e-05,
|
| 692 |
+
"loss": 0.2906,
|
| 693 |
+
"step": 990
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 10.42,
|
| 697 |
+
"learning_rate": 4.1087962962962965e-05,
|
| 698 |
+
"loss": 0.3518,
|
| 699 |
+
"step": 1000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 10.52,
|
| 703 |
+
"learning_rate": 4.094328703703704e-05,
|
| 704 |
+
"loss": 0.3341,
|
| 705 |
+
"step": 1010
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 10.62,
|
| 709 |
+
"learning_rate": 4.0798611111111115e-05,
|
| 710 |
+
"loss": 0.3725,
|
| 711 |
+
"step": 1020
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 10.73,
|
| 715 |
+
"learning_rate": 4.065393518518519e-05,
|
| 716 |
+
"loss": 0.3435,
|
| 717 |
+
"step": 1030
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 10.83,
|
| 721 |
+
"learning_rate": 4.0509259259259265e-05,
|
| 722 |
+
"loss": 0.3219,
|
| 723 |
+
"step": 1040
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 10.94,
|
| 727 |
+
"learning_rate": 4.036458333333333e-05,
|
| 728 |
+
"loss": 0.3301,
|
| 729 |
+
"step": 1050
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 11.0,
|
| 733 |
+
"eval_accuracy": 0.5581818181818182,
|
| 734 |
+
"eval_loss": 2.7200424671173096,
|
| 735 |
+
"eval_runtime": 3.7269,
|
| 736 |
+
"eval_samples_per_second": 147.576,
|
| 737 |
+
"eval_steps_per_second": 9.391,
|
| 738 |
+
"step": 1056
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 11.04,
|
| 742 |
+
"learning_rate": 4.021990740740741e-05,
|
| 743 |
+
"loss": 0.3121,
|
| 744 |
+
"step": 1060
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 11.15,
|
| 748 |
+
"learning_rate": 4.007523148148148e-05,
|
| 749 |
+
"loss": 0.3416,
|
| 750 |
+
"step": 1070
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 11.25,
|
| 754 |
+
"learning_rate": 3.993055555555556e-05,
|
| 755 |
+
"loss": 0.2896,
|
| 756 |
+
"step": 1080
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 11.35,
|
| 760 |
+
"learning_rate": 3.978587962962963e-05,
|
| 761 |
+
"loss": 0.3542,
|
| 762 |
+
"step": 1090
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 11.46,
|
| 766 |
+
"learning_rate": 3.964120370370371e-05,
|
| 767 |
+
"loss": 0.3516,
|
| 768 |
+
"step": 1100
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 11.56,
|
| 772 |
+
"learning_rate": 3.949652777777778e-05,
|
| 773 |
+
"loss": 0.295,
|
| 774 |
+
"step": 1110
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 11.67,
|
| 778 |
+
"learning_rate": 3.935185185185186e-05,
|
| 779 |
+
"loss": 0.3252,
|
| 780 |
+
"step": 1120
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 11.77,
|
| 784 |
+
"learning_rate": 3.9207175925925926e-05,
|
| 785 |
+
"loss": 0.2624,
|
| 786 |
+
"step": 1130
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 11.88,
|
| 790 |
+
"learning_rate": 3.90625e-05,
|
| 791 |
+
"loss": 0.329,
|
| 792 |
+
"step": 1140
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 11.98,
|
| 796 |
+
"learning_rate": 3.8917824074074076e-05,
|
| 797 |
+
"loss": 0.2786,
|
| 798 |
+
"step": 1150
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 12.0,
|
| 802 |
+
"eval_accuracy": 0.6272727272727273,
|
| 803 |
+
"eval_loss": 1.8999561071395874,
|
| 804 |
+
"eval_runtime": 3.6939,
|
| 805 |
+
"eval_samples_per_second": 148.893,
|
| 806 |
+
"eval_steps_per_second": 9.475,
|
| 807 |
+
"step": 1152
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 12.08,
|
| 811 |
+
"learning_rate": 3.877314814814815e-05,
|
| 812 |
+
"loss": 0.2833,
|
| 813 |
+
"step": 1160
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 12.19,
|
| 817 |
+
"learning_rate": 3.8628472222222226e-05,
|
| 818 |
+
"loss": 0.2536,
|
| 819 |
+
"step": 1170
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 12.29,
|
| 823 |
+
"learning_rate": 3.84837962962963e-05,
|
| 824 |
+
"loss": 0.3003,
|
| 825 |
+
"step": 1180
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 12.4,
|
| 829 |
+
"learning_rate": 3.8339120370370376e-05,
|
| 830 |
+
"loss": 0.2713,
|
| 831 |
+
"step": 1190
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 12.5,
|
| 835 |
+
"learning_rate": 3.8194444444444444e-05,
|
| 836 |
+
"loss": 0.2854,
|
| 837 |
+
"step": 1200
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 12.6,
|
| 841 |
+
"learning_rate": 3.804976851851852e-05,
|
| 842 |
+
"loss": 0.2695,
|
| 843 |
+
"step": 1210
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 12.71,
|
| 847 |
+
"learning_rate": 3.7905092592592594e-05,
|
| 848 |
+
"loss": 0.2702,
|
| 849 |
+
"step": 1220
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 12.81,
|
| 853 |
+
"learning_rate": 3.776041666666667e-05,
|
| 854 |
+
"loss": 0.3312,
|
| 855 |
+
"step": 1230
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 12.92,
|
| 859 |
+
"learning_rate": 3.7615740740740744e-05,
|
| 860 |
+
"loss": 0.2746,
|
| 861 |
+
"step": 1240
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 13.0,
|
| 865 |
+
"eval_accuracy": 0.5363636363636364,
|
| 866 |
+
"eval_loss": 3.1768321990966797,
|
| 867 |
+
"eval_runtime": 3.7274,
|
| 868 |
+
"eval_samples_per_second": 147.555,
|
| 869 |
+
"eval_steps_per_second": 9.39,
|
| 870 |
+
"step": 1248
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 13.02,
|
| 874 |
+
"learning_rate": 3.747106481481482e-05,
|
| 875 |
+
"loss": 0.2632,
|
| 876 |
+
"step": 1250
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 13.12,
|
| 880 |
+
"learning_rate": 3.7326388888888893e-05,
|
| 881 |
+
"loss": 0.2495,
|
| 882 |
+
"step": 1260
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 13.23,
|
| 886 |
+
"learning_rate": 3.718171296296297e-05,
|
| 887 |
+
"loss": 0.2761,
|
| 888 |
+
"step": 1270
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 13.33,
|
| 892 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 893 |
+
"loss": 0.2879,
|
| 894 |
+
"step": 1280
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 13.44,
|
| 898 |
+
"learning_rate": 3.689236111111111e-05,
|
| 899 |
+
"loss": 0.2658,
|
| 900 |
+
"step": 1290
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 13.54,
|
| 904 |
+
"learning_rate": 3.6747685185185186e-05,
|
| 905 |
+
"loss": 0.2652,
|
| 906 |
+
"step": 1300
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 13.65,
|
| 910 |
+
"learning_rate": 3.660300925925926e-05,
|
| 911 |
+
"loss": 0.2604,
|
| 912 |
+
"step": 1310
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 13.75,
|
| 916 |
+
"learning_rate": 3.6458333333333336e-05,
|
| 917 |
+
"loss": 0.2468,
|
| 918 |
+
"step": 1320
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 13.85,
|
| 922 |
+
"learning_rate": 3.631365740740741e-05,
|
| 923 |
+
"loss": 0.2552,
|
| 924 |
+
"step": 1330
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 13.96,
|
| 928 |
+
"learning_rate": 3.6168981481481486e-05,
|
| 929 |
+
"loss": 0.2298,
|
| 930 |
+
"step": 1340
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 14.0,
|
| 934 |
+
"eval_accuracy": 0.5527272727272727,
|
| 935 |
+
"eval_loss": 3.100266218185425,
|
| 936 |
+
"eval_runtime": 3.7819,
|
| 937 |
+
"eval_samples_per_second": 145.429,
|
| 938 |
+
"eval_steps_per_second": 9.255,
|
| 939 |
+
"step": 1344
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 14.06,
|
| 943 |
+
"learning_rate": 3.602430555555556e-05,
|
| 944 |
+
"loss": 0.2423,
|
| 945 |
+
"step": 1350
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 14.17,
|
| 949 |
+
"learning_rate": 3.587962962962963e-05,
|
| 950 |
+
"loss": 0.2102,
|
| 951 |
+
"step": 1360
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 14.27,
|
| 955 |
+
"learning_rate": 3.5734953703703704e-05,
|
| 956 |
+
"loss": 0.207,
|
| 957 |
+
"step": 1370
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 14.38,
|
| 961 |
+
"learning_rate": 3.559027777777778e-05,
|
| 962 |
+
"loss": 0.2548,
|
| 963 |
+
"step": 1380
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 14.48,
|
| 967 |
+
"learning_rate": 3.5445601851851854e-05,
|
| 968 |
+
"loss": 0.2428,
|
| 969 |
+
"step": 1390
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 14.58,
|
| 973 |
+
"learning_rate": 3.530092592592593e-05,
|
| 974 |
+
"loss": 0.2344,
|
| 975 |
+
"step": 1400
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 14.69,
|
| 979 |
+
"learning_rate": 3.5156250000000004e-05,
|
| 980 |
+
"loss": 0.2847,
|
| 981 |
+
"step": 1410
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 14.79,
|
| 985 |
+
"learning_rate": 3.501157407407408e-05,
|
| 986 |
+
"loss": 0.2296,
|
| 987 |
+
"step": 1420
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 14.9,
|
| 991 |
+
"learning_rate": 3.486689814814815e-05,
|
| 992 |
+
"loss": 0.2894,
|
| 993 |
+
"step": 1430
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 15.0,
|
| 997 |
+
"learning_rate": 3.472222222222222e-05,
|
| 998 |
+
"loss": 0.2013,
|
| 999 |
+
"step": 1440
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 15.0,
|
| 1003 |
+
"eval_accuracy": 0.6181818181818182,
|
| 1004 |
+
"eval_loss": 2.344068765640259,
|
| 1005 |
+
"eval_runtime": 3.8014,
|
| 1006 |
+
"eval_samples_per_second": 144.683,
|
| 1007 |
+
"eval_steps_per_second": 9.207,
|
| 1008 |
+
"step": 1440
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 15.1,
|
| 1012 |
+
"learning_rate": 3.45775462962963e-05,
|
| 1013 |
+
"loss": 0.2093,
|
| 1014 |
+
"step": 1450
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 15.21,
|
| 1018 |
+
"learning_rate": 3.443287037037037e-05,
|
| 1019 |
+
"loss": 0.2092,
|
| 1020 |
+
"step": 1460
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 15.31,
|
| 1024 |
+
"learning_rate": 3.428819444444444e-05,
|
| 1025 |
+
"loss": 0.1995,
|
| 1026 |
+
"step": 1470
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 15.42,
|
| 1030 |
+
"learning_rate": 3.414351851851852e-05,
|
| 1031 |
+
"loss": 0.2162,
|
| 1032 |
+
"step": 1480
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 15.52,
|
| 1036 |
+
"learning_rate": 3.39988425925926e-05,
|
| 1037 |
+
"loss": 0.2444,
|
| 1038 |
+
"step": 1490
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 15.62,
|
| 1042 |
+
"learning_rate": 3.385416666666667e-05,
|
| 1043 |
+
"loss": 0.1702,
|
| 1044 |
+
"step": 1500
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 15.73,
|
| 1048 |
+
"learning_rate": 3.370949074074074e-05,
|
| 1049 |
+
"loss": 0.28,
|
| 1050 |
+
"step": 1510
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 15.83,
|
| 1054 |
+
"learning_rate": 3.3564814814814815e-05,
|
| 1055 |
+
"loss": 0.2303,
|
| 1056 |
+
"step": 1520
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 15.94,
|
| 1060 |
+
"learning_rate": 3.342013888888889e-05,
|
| 1061 |
+
"loss": 0.2225,
|
| 1062 |
+
"step": 1530
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 16.0,
|
| 1066 |
+
"eval_accuracy": 0.5709090909090909,
|
| 1067 |
+
"eval_loss": 3.0213537216186523,
|
| 1068 |
+
"eval_runtime": 3.7244,
|
| 1069 |
+
"eval_samples_per_second": 147.674,
|
| 1070 |
+
"eval_steps_per_second": 9.397,
|
| 1071 |
+
"step": 1536
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 16.04,
|
| 1075 |
+
"learning_rate": 3.3275462962962965e-05,
|
| 1076 |
+
"loss": 0.2012,
|
| 1077 |
+
"step": 1540
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 16.15,
|
| 1081 |
+
"learning_rate": 3.313078703703704e-05,
|
| 1082 |
+
"loss": 0.1761,
|
| 1083 |
+
"step": 1550
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 16.25,
|
| 1087 |
+
"learning_rate": 3.2986111111111115e-05,
|
| 1088 |
+
"loss": 0.229,
|
| 1089 |
+
"step": 1560
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 16.35,
|
| 1093 |
+
"learning_rate": 3.284143518518519e-05,
|
| 1094 |
+
"loss": 0.2114,
|
| 1095 |
+
"step": 1570
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 16.46,
|
| 1099 |
+
"learning_rate": 3.2696759259259265e-05,
|
| 1100 |
+
"loss": 0.2284,
|
| 1101 |
+
"step": 1580
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 16.56,
|
| 1105 |
+
"learning_rate": 3.255208333333333e-05,
|
| 1106 |
+
"loss": 0.2137,
|
| 1107 |
+
"step": 1590
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 16.67,
|
| 1111 |
+
"learning_rate": 3.240740740740741e-05,
|
| 1112 |
+
"loss": 0.2902,
|
| 1113 |
+
"step": 1600
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 16.77,
|
| 1117 |
+
"learning_rate": 3.226273148148148e-05,
|
| 1118 |
+
"loss": 0.2273,
|
| 1119 |
+
"step": 1610
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 16.88,
|
| 1123 |
+
"learning_rate": 3.211805555555556e-05,
|
| 1124 |
+
"loss": 0.182,
|
| 1125 |
+
"step": 1620
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 16.98,
|
| 1129 |
+
"learning_rate": 3.197337962962963e-05,
|
| 1130 |
+
"loss": 0.2229,
|
| 1131 |
+
"step": 1630
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 17.0,
|
| 1135 |
+
"eval_accuracy": 0.6163636363636363,
|
| 1136 |
+
"eval_loss": 2.067636728286743,
|
| 1137 |
+
"eval_runtime": 3.8554,
|
| 1138 |
+
"eval_samples_per_second": 142.656,
|
| 1139 |
+
"eval_steps_per_second": 9.078,
|
| 1140 |
+
"step": 1632
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 17.08,
|
| 1144 |
+
"learning_rate": 3.182870370370371e-05,
|
| 1145 |
+
"loss": 0.1704,
|
| 1146 |
+
"step": 1640
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 17.19,
|
| 1150 |
+
"learning_rate": 3.168402777777778e-05,
|
| 1151 |
+
"loss": 0.2184,
|
| 1152 |
+
"step": 1650
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 17.29,
|
| 1156 |
+
"learning_rate": 3.153935185185186e-05,
|
| 1157 |
+
"loss": 0.1662,
|
| 1158 |
+
"step": 1660
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 17.4,
|
| 1162 |
+
"learning_rate": 3.1394675925925926e-05,
|
| 1163 |
+
"loss": 0.1965,
|
| 1164 |
+
"step": 1670
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 17.5,
|
| 1168 |
+
"learning_rate": 3.125e-05,
|
| 1169 |
+
"loss": 0.1384,
|
| 1170 |
+
"step": 1680
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 17.6,
|
| 1174 |
+
"learning_rate": 3.1105324074074076e-05,
|
| 1175 |
+
"loss": 0.213,
|
| 1176 |
+
"step": 1690
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 17.71,
|
| 1180 |
+
"learning_rate": 3.0960648148148144e-05,
|
| 1181 |
+
"loss": 0.2053,
|
| 1182 |
+
"step": 1700
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 17.81,
|
| 1186 |
+
"learning_rate": 3.0815972222222225e-05,
|
| 1187 |
+
"loss": 0.1938,
|
| 1188 |
+
"step": 1710
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 17.92,
|
| 1192 |
+
"learning_rate": 3.06712962962963e-05,
|
| 1193 |
+
"loss": 0.2024,
|
| 1194 |
+
"step": 1720
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 18.0,
|
| 1198 |
+
"eval_accuracy": 0.5672727272727273,
|
| 1199 |
+
"eval_loss": 2.6478159427642822,
|
| 1200 |
+
"eval_runtime": 3.8359,
|
| 1201 |
+
"eval_samples_per_second": 143.382,
|
| 1202 |
+
"eval_steps_per_second": 9.124,
|
| 1203 |
+
"step": 1728
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 18.02,
|
| 1207 |
+
"learning_rate": 3.0526620370370375e-05,
|
| 1208 |
+
"loss": 0.1643,
|
| 1209 |
+
"step": 1730
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 18.12,
|
| 1213 |
+
"learning_rate": 3.0381944444444444e-05,
|
| 1214 |
+
"loss": 0.2376,
|
| 1215 |
+
"step": 1740
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 18.23,
|
| 1219 |
+
"learning_rate": 3.023726851851852e-05,
|
| 1220 |
+
"loss": 0.2088,
|
| 1221 |
+
"step": 1750
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 18.33,
|
| 1225 |
+
"learning_rate": 3.0092592592592593e-05,
|
| 1226 |
+
"loss": 0.181,
|
| 1227 |
+
"step": 1760
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 18.44,
|
| 1231 |
+
"learning_rate": 2.994791666666667e-05,
|
| 1232 |
+
"loss": 0.19,
|
| 1233 |
+
"step": 1770
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 18.54,
|
| 1237 |
+
"learning_rate": 2.980324074074074e-05,
|
| 1238 |
+
"loss": 0.2008,
|
| 1239 |
+
"step": 1780
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 18.65,
|
| 1243 |
+
"learning_rate": 2.9658564814814815e-05,
|
| 1244 |
+
"loss": 0.2109,
|
| 1245 |
+
"step": 1790
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 18.75,
|
| 1249 |
+
"learning_rate": 2.951388888888889e-05,
|
| 1250 |
+
"loss": 0.1904,
|
| 1251 |
+
"step": 1800
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 18.85,
|
| 1255 |
+
"learning_rate": 2.9369212962962965e-05,
|
| 1256 |
+
"loss": 0.2136,
|
| 1257 |
+
"step": 1810
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 18.96,
|
| 1261 |
+
"learning_rate": 2.9224537037037036e-05,
|
| 1262 |
+
"loss": 0.1401,
|
| 1263 |
+
"step": 1820
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 19.0,
|
| 1267 |
+
"eval_accuracy": 0.5636363636363636,
|
| 1268 |
+
"eval_loss": 2.8951961994171143,
|
| 1269 |
+
"eval_runtime": 3.8754,
|
| 1270 |
+
"eval_samples_per_second": 141.92,
|
| 1271 |
+
"eval_steps_per_second": 9.031,
|
| 1272 |
+
"step": 1824
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 19.06,
|
| 1276 |
+
"learning_rate": 2.907986111111111e-05,
|
| 1277 |
+
"loss": 0.1974,
|
| 1278 |
+
"step": 1830
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 19.17,
|
| 1282 |
+
"learning_rate": 2.8935185185185186e-05,
|
| 1283 |
+
"loss": 0.1752,
|
| 1284 |
+
"step": 1840
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 19.27,
|
| 1288 |
+
"learning_rate": 2.879050925925926e-05,
|
| 1289 |
+
"loss": 0.1814,
|
| 1290 |
+
"step": 1850
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 19.38,
|
| 1294 |
+
"learning_rate": 2.8645833333333333e-05,
|
| 1295 |
+
"loss": 0.1853,
|
| 1296 |
+
"step": 1860
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 19.48,
|
| 1300 |
+
"learning_rate": 2.8501157407407408e-05,
|
| 1301 |
+
"loss": 0.1992,
|
| 1302 |
+
"step": 1870
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 19.58,
|
| 1306 |
+
"learning_rate": 2.8356481481481483e-05,
|
| 1307 |
+
"loss": 0.1774,
|
| 1308 |
+
"step": 1880
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 19.69,
|
| 1312 |
+
"learning_rate": 2.821180555555556e-05,
|
| 1313 |
+
"loss": 0.1836,
|
| 1314 |
+
"step": 1890
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 19.79,
|
| 1318 |
+
"learning_rate": 2.806712962962963e-05,
|
| 1319 |
+
"loss": 0.1556,
|
| 1320 |
+
"step": 1900
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 19.9,
|
| 1324 |
+
"learning_rate": 2.7922453703703704e-05,
|
| 1325 |
+
"loss": 0.175,
|
| 1326 |
+
"step": 1910
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 20.0,
|
| 1330 |
+
"learning_rate": 2.777777777777778e-05,
|
| 1331 |
+
"loss": 0.1984,
|
| 1332 |
+
"step": 1920
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 20.0,
|
| 1336 |
+
"eval_accuracy": 0.6145454545454545,
|
| 1337 |
+
"eval_loss": 2.308309555053711,
|
| 1338 |
+
"eval_runtime": 3.9255,
|
| 1339 |
+
"eval_samples_per_second": 140.111,
|
| 1340 |
+
"eval_steps_per_second": 8.916,
|
| 1341 |
+
"step": 1920
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 20.1,
|
| 1345 |
+
"learning_rate": 2.7633101851851857e-05,
|
| 1346 |
+
"loss": 0.1598,
|
| 1347 |
+
"step": 1930
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 20.21,
|
| 1351 |
+
"learning_rate": 2.7488425925925926e-05,
|
| 1352 |
+
"loss": 0.1761,
|
| 1353 |
+
"step": 1940
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 20.31,
|
| 1357 |
+
"learning_rate": 2.734375e-05,
|
| 1358 |
+
"loss": 0.1647,
|
| 1359 |
+
"step": 1950
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 20.42,
|
| 1363 |
+
"learning_rate": 2.7199074074074076e-05,
|
| 1364 |
+
"loss": 0.1553,
|
| 1365 |
+
"step": 1960
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 20.52,
|
| 1369 |
+
"learning_rate": 2.7054398148148147e-05,
|
| 1370 |
+
"loss": 0.1591,
|
| 1371 |
+
"step": 1970
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 20.62,
|
| 1375 |
+
"learning_rate": 2.6909722222222222e-05,
|
| 1376 |
+
"loss": 0.1736,
|
| 1377 |
+
"step": 1980
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 20.73,
|
| 1381 |
+
"learning_rate": 2.6765046296296297e-05,
|
| 1382 |
+
"loss": 0.1892,
|
| 1383 |
+
"step": 1990
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 20.83,
|
| 1387 |
+
"learning_rate": 2.6620370370370372e-05,
|
| 1388 |
+
"loss": 0.1596,
|
| 1389 |
+
"step": 2000
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 20.94,
|
| 1393 |
+
"learning_rate": 2.6475694444444443e-05,
|
| 1394 |
+
"loss": 0.1788,
|
| 1395 |
+
"step": 2010
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 21.0,
|
| 1399 |
+
"eval_accuracy": 0.52,
|
| 1400 |
+
"eval_loss": 3.7701735496520996,
|
| 1401 |
+
"eval_runtime": 3.9464,
|
| 1402 |
+
"eval_samples_per_second": 139.366,
|
| 1403 |
+
"eval_steps_per_second": 8.869,
|
| 1404 |
+
"step": 2016
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 21.04,
|
| 1408 |
+
"learning_rate": 2.633101851851852e-05,
|
| 1409 |
+
"loss": 0.1783,
|
| 1410 |
+
"step": 2020
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 21.15,
|
| 1414 |
+
"learning_rate": 2.6186342592592593e-05,
|
| 1415 |
+
"loss": 0.1855,
|
| 1416 |
+
"step": 2030
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 21.25,
|
| 1420 |
+
"learning_rate": 2.604166666666667e-05,
|
| 1421 |
+
"loss": 0.2003,
|
| 1422 |
+
"step": 2040
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 21.35,
|
| 1426 |
+
"learning_rate": 2.589699074074074e-05,
|
| 1427 |
+
"loss": 0.1726,
|
| 1428 |
+
"step": 2050
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 21.46,
|
| 1432 |
+
"learning_rate": 2.5752314814814815e-05,
|
| 1433 |
+
"loss": 0.153,
|
| 1434 |
+
"step": 2060
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 21.56,
|
| 1438 |
+
"learning_rate": 2.560763888888889e-05,
|
| 1439 |
+
"loss": 0.1474,
|
| 1440 |
+
"step": 2070
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 21.67,
|
| 1444 |
+
"learning_rate": 2.5462962962962965e-05,
|
| 1445 |
+
"loss": 0.1354,
|
| 1446 |
+
"step": 2080
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 21.77,
|
| 1450 |
+
"learning_rate": 2.5318287037037036e-05,
|
| 1451 |
+
"loss": 0.1553,
|
| 1452 |
+
"step": 2090
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 21.88,
|
| 1456 |
+
"learning_rate": 2.517361111111111e-05,
|
| 1457 |
+
"loss": 0.1832,
|
| 1458 |
+
"step": 2100
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 21.98,
|
| 1462 |
+
"learning_rate": 2.5028935185185186e-05,
|
| 1463 |
+
"loss": 0.1907,
|
| 1464 |
+
"step": 2110
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 22.0,
|
| 1468 |
+
"eval_accuracy": 0.6654545454545454,
|
| 1469 |
+
"eval_loss": 1.9616814851760864,
|
| 1470 |
+
"eval_runtime": 3.8632,
|
| 1471 |
+
"eval_samples_per_second": 142.368,
|
| 1472 |
+
"eval_steps_per_second": 9.06,
|
| 1473 |
+
"step": 2112
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 22.08,
|
| 1477 |
+
"learning_rate": 2.488425925925926e-05,
|
| 1478 |
+
"loss": 0.159,
|
| 1479 |
+
"step": 2120
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 22.19,
|
| 1483 |
+
"learning_rate": 2.4739583333333336e-05,
|
| 1484 |
+
"loss": 0.1397,
|
| 1485 |
+
"step": 2130
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 22.29,
|
| 1489 |
+
"learning_rate": 2.4594907407407408e-05,
|
| 1490 |
+
"loss": 0.1511,
|
| 1491 |
+
"step": 2140
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 22.4,
|
| 1495 |
+
"learning_rate": 2.4450231481481483e-05,
|
| 1496 |
+
"loss": 0.1704,
|
| 1497 |
+
"step": 2150
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 22.5,
|
| 1501 |
+
"learning_rate": 2.4305555555555558e-05,
|
| 1502 |
+
"loss": 0.1475,
|
| 1503 |
+
"step": 2160
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 22.6,
|
| 1507 |
+
"learning_rate": 2.4160879629629633e-05,
|
| 1508 |
+
"loss": 0.1568,
|
| 1509 |
+
"step": 2170
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 22.71,
|
| 1513 |
+
"learning_rate": 2.4016203703703704e-05,
|
| 1514 |
+
"loss": 0.1394,
|
| 1515 |
+
"step": 2180
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 22.81,
|
| 1519 |
+
"learning_rate": 2.387152777777778e-05,
|
| 1520 |
+
"loss": 0.1274,
|
| 1521 |
+
"step": 2190
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 22.92,
|
| 1525 |
+
"learning_rate": 2.3726851851851854e-05,
|
| 1526 |
+
"loss": 0.1113,
|
| 1527 |
+
"step": 2200
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 23.0,
|
| 1531 |
+
"eval_accuracy": 0.5963636363636363,
|
| 1532 |
+
"eval_loss": 2.654604434967041,
|
| 1533 |
+
"eval_runtime": 3.8399,
|
| 1534 |
+
"eval_samples_per_second": 143.233,
|
| 1535 |
+
"eval_steps_per_second": 9.115,
|
| 1536 |
+
"step": 2208
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 23.02,
|
| 1540 |
+
"learning_rate": 2.358217592592593e-05,
|
| 1541 |
+
"loss": 0.1994,
|
| 1542 |
+
"step": 2210
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 23.12,
|
| 1546 |
+
"learning_rate": 2.34375e-05,
|
| 1547 |
+
"loss": 0.1513,
|
| 1548 |
+
"step": 2220
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 23.23,
|
| 1552 |
+
"learning_rate": 2.3292824074074075e-05,
|
| 1553 |
+
"loss": 0.1596,
|
| 1554 |
+
"step": 2230
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 23.33,
|
| 1558 |
+
"learning_rate": 2.314814814814815e-05,
|
| 1559 |
+
"loss": 0.1276,
|
| 1560 |
+
"step": 2240
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 23.44,
|
| 1564 |
+
"learning_rate": 2.3003472222222222e-05,
|
| 1565 |
+
"loss": 0.1552,
|
| 1566 |
+
"step": 2250
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 23.54,
|
| 1570 |
+
"learning_rate": 2.2858796296296297e-05,
|
| 1571 |
+
"loss": 0.1419,
|
| 1572 |
+
"step": 2260
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 23.65,
|
| 1576 |
+
"learning_rate": 2.2714120370370372e-05,
|
| 1577 |
+
"loss": 0.1798,
|
| 1578 |
+
"step": 2270
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 23.75,
|
| 1582 |
+
"learning_rate": 2.2569444444444447e-05,
|
| 1583 |
+
"loss": 0.1832,
|
| 1584 |
+
"step": 2280
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 23.85,
|
| 1588 |
+
"learning_rate": 2.242476851851852e-05,
|
| 1589 |
+
"loss": 0.1656,
|
| 1590 |
+
"step": 2290
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 23.96,
|
| 1594 |
+
"learning_rate": 2.2280092592592593e-05,
|
| 1595 |
+
"loss": 0.1293,
|
| 1596 |
+
"step": 2300
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 24.0,
|
| 1600 |
+
"eval_accuracy": 0.6036363636363636,
|
| 1601 |
+
"eval_loss": 2.6427228450775146,
|
| 1602 |
+
"eval_runtime": 3.9185,
|
| 1603 |
+
"eval_samples_per_second": 140.361,
|
| 1604 |
+
"eval_steps_per_second": 8.932,
|
| 1605 |
+
"step": 2304
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 24.06,
|
| 1609 |
+
"learning_rate": 2.2135416666666668e-05,
|
| 1610 |
+
"loss": 0.1278,
|
| 1611 |
+
"step": 2310
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 24.17,
|
| 1615 |
+
"learning_rate": 2.1990740740740743e-05,
|
| 1616 |
+
"loss": 0.1154,
|
| 1617 |
+
"step": 2320
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 24.27,
|
| 1621 |
+
"learning_rate": 2.1846064814814815e-05,
|
| 1622 |
+
"loss": 0.1542,
|
| 1623 |
+
"step": 2330
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 24.38,
|
| 1627 |
+
"learning_rate": 2.170138888888889e-05,
|
| 1628 |
+
"loss": 0.2238,
|
| 1629 |
+
"step": 2340
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 24.48,
|
| 1633 |
+
"learning_rate": 2.1556712962962965e-05,
|
| 1634 |
+
"loss": 0.16,
|
| 1635 |
+
"step": 2350
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 24.58,
|
| 1639 |
+
"learning_rate": 2.141203703703704e-05,
|
| 1640 |
+
"loss": 0.1713,
|
| 1641 |
+
"step": 2360
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 24.69,
|
| 1645 |
+
"learning_rate": 2.126736111111111e-05,
|
| 1646 |
+
"loss": 0.1392,
|
| 1647 |
+
"step": 2370
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 24.79,
|
| 1651 |
+
"learning_rate": 2.1122685185185186e-05,
|
| 1652 |
+
"loss": 0.1586,
|
| 1653 |
+
"step": 2380
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 24.9,
|
| 1657 |
+
"learning_rate": 2.097800925925926e-05,
|
| 1658 |
+
"loss": 0.0984,
|
| 1659 |
+
"step": 2390
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 25.0,
|
| 1663 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 1664 |
+
"loss": 0.1354,
|
| 1665 |
+
"step": 2400
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 25.0,
|
| 1669 |
+
"eval_accuracy": 0.5527272727272727,
|
| 1670 |
+
"eval_loss": 3.410457134246826,
|
| 1671 |
+
"eval_runtime": 3.9002,
|
| 1672 |
+
"eval_samples_per_second": 141.019,
|
| 1673 |
+
"eval_steps_per_second": 8.974,
|
| 1674 |
+
"step": 2400
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 25.1,
|
| 1678 |
+
"learning_rate": 2.0688657407407408e-05,
|
| 1679 |
+
"loss": 0.1183,
|
| 1680 |
+
"step": 2410
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 25.21,
|
| 1684 |
+
"learning_rate": 2.0543981481481483e-05,
|
| 1685 |
+
"loss": 0.1221,
|
| 1686 |
+
"step": 2420
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 25.31,
|
| 1690 |
+
"learning_rate": 2.0399305555555557e-05,
|
| 1691 |
+
"loss": 0.1588,
|
| 1692 |
+
"step": 2430
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 25.42,
|
| 1696 |
+
"learning_rate": 2.0254629629629632e-05,
|
| 1697 |
+
"loss": 0.1581,
|
| 1698 |
+
"step": 2440
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 25.52,
|
| 1702 |
+
"learning_rate": 2.0109953703703704e-05,
|
| 1703 |
+
"loss": 0.1372,
|
| 1704 |
+
"step": 2450
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 25.62,
|
| 1708 |
+
"learning_rate": 1.996527777777778e-05,
|
| 1709 |
+
"loss": 0.1165,
|
| 1710 |
+
"step": 2460
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 25.73,
|
| 1714 |
+
"learning_rate": 1.9820601851851854e-05,
|
| 1715 |
+
"loss": 0.124,
|
| 1716 |
+
"step": 2470
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 25.83,
|
| 1720 |
+
"learning_rate": 1.967592592592593e-05,
|
| 1721 |
+
"loss": 0.1127,
|
| 1722 |
+
"step": 2480
|
| 1723 |
+
},
|
| 1724 |
+
{
|
| 1725 |
+
"epoch": 25.94,
|
| 1726 |
+
"learning_rate": 1.953125e-05,
|
| 1727 |
+
"loss": 0.1447,
|
| 1728 |
+
"step": 2490
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 26.0,
|
| 1732 |
+
"eval_accuracy": 0.6127272727272727,
|
| 1733 |
+
"eval_loss": 2.545973062515259,
|
| 1734 |
+
"eval_runtime": 4.0556,
|
| 1735 |
+
"eval_samples_per_second": 135.614,
|
| 1736 |
+
"eval_steps_per_second": 8.63,
|
| 1737 |
+
"step": 2496
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 26.04,
|
| 1741 |
+
"learning_rate": 1.9386574074074075e-05,
|
| 1742 |
+
"loss": 0.1788,
|
| 1743 |
+
"step": 2500
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 26.15,
|
| 1747 |
+
"learning_rate": 1.924189814814815e-05,
|
| 1748 |
+
"loss": 0.1482,
|
| 1749 |
+
"step": 2510
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 26.25,
|
| 1753 |
+
"learning_rate": 1.9097222222222222e-05,
|
| 1754 |
+
"loss": 0.126,
|
| 1755 |
+
"step": 2520
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 26.35,
|
| 1759 |
+
"learning_rate": 1.8952546296296297e-05,
|
| 1760 |
+
"loss": 0.1427,
|
| 1761 |
+
"step": 2530
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 26.46,
|
| 1765 |
+
"learning_rate": 1.8807870370370372e-05,
|
| 1766 |
+
"loss": 0.1603,
|
| 1767 |
+
"step": 2540
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 26.56,
|
| 1771 |
+
"learning_rate": 1.8663194444444447e-05,
|
| 1772 |
+
"loss": 0.1391,
|
| 1773 |
+
"step": 2550
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 26.67,
|
| 1777 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 1778 |
+
"loss": 0.1036,
|
| 1779 |
+
"step": 2560
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 26.77,
|
| 1783 |
+
"learning_rate": 1.8373842592592593e-05,
|
| 1784 |
+
"loss": 0.1336,
|
| 1785 |
+
"step": 2570
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 26.88,
|
| 1789 |
+
"learning_rate": 1.8229166666666668e-05,
|
| 1790 |
+
"loss": 0.1348,
|
| 1791 |
+
"step": 2580
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 26.98,
|
| 1795 |
+
"learning_rate": 1.8084490740740743e-05,
|
| 1796 |
+
"loss": 0.0995,
|
| 1797 |
+
"step": 2590
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 27.0,
|
| 1801 |
+
"eval_accuracy": 0.5854545454545454,
|
| 1802 |
+
"eval_loss": 2.9865081310272217,
|
| 1803 |
+
"eval_runtime": 3.9958,
|
| 1804 |
+
"eval_samples_per_second": 137.646,
|
| 1805 |
+
"eval_steps_per_second": 8.759,
|
| 1806 |
+
"step": 2592
|
| 1807 |
+
},
|
| 1808 |
+
{
|
| 1809 |
+
"epoch": 27.08,
|
| 1810 |
+
"learning_rate": 1.7939814814814815e-05,
|
| 1811 |
+
"loss": 0.1093,
|
| 1812 |
+
"step": 2600
|
| 1813 |
+
},
|
| 1814 |
+
{
|
| 1815 |
+
"epoch": 27.19,
|
| 1816 |
+
"learning_rate": 1.779513888888889e-05,
|
| 1817 |
+
"loss": 0.1454,
|
| 1818 |
+
"step": 2610
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 27.29,
|
| 1822 |
+
"learning_rate": 1.7650462962962965e-05,
|
| 1823 |
+
"loss": 0.1408,
|
| 1824 |
+
"step": 2620
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 27.4,
|
| 1828 |
+
"learning_rate": 1.750578703703704e-05,
|
| 1829 |
+
"loss": 0.1415,
|
| 1830 |
+
"step": 2630
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 27.5,
|
| 1834 |
+
"learning_rate": 1.736111111111111e-05,
|
| 1835 |
+
"loss": 0.1032,
|
| 1836 |
+
"step": 2640
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 27.6,
|
| 1840 |
+
"learning_rate": 1.7216435185185186e-05,
|
| 1841 |
+
"loss": 0.1551,
|
| 1842 |
+
"step": 2650
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 27.71,
|
| 1846 |
+
"learning_rate": 1.707175925925926e-05,
|
| 1847 |
+
"loss": 0.084,
|
| 1848 |
+
"step": 2660
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"epoch": 27.81,
|
| 1852 |
+
"learning_rate": 1.6927083333333336e-05,
|
| 1853 |
+
"loss": 0.1088,
|
| 1854 |
+
"step": 2670
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 27.92,
|
| 1858 |
+
"learning_rate": 1.6782407407407408e-05,
|
| 1859 |
+
"loss": 0.1369,
|
| 1860 |
+
"step": 2680
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 28.0,
|
| 1864 |
+
"eval_accuracy": 0.5545454545454546,
|
| 1865 |
+
"eval_loss": 3.5280685424804688,
|
| 1866 |
+
"eval_runtime": 3.88,
|
| 1867 |
+
"eval_samples_per_second": 141.752,
|
| 1868 |
+
"eval_steps_per_second": 9.021,
|
| 1869 |
+
"step": 2688
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 28.02,
|
| 1873 |
+
"learning_rate": 1.6637731481481482e-05,
|
| 1874 |
+
"loss": 0.14,
|
| 1875 |
+
"step": 2690
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 28.12,
|
| 1879 |
+
"learning_rate": 1.6493055555555557e-05,
|
| 1880 |
+
"loss": 0.1198,
|
| 1881 |
+
"step": 2700
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 28.23,
|
| 1885 |
+
"learning_rate": 1.6348379629629632e-05,
|
| 1886 |
+
"loss": 0.1132,
|
| 1887 |
+
"step": 2710
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 28.33,
|
| 1891 |
+
"learning_rate": 1.6203703703703704e-05,
|
| 1892 |
+
"loss": 0.1526,
|
| 1893 |
+
"step": 2720
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 28.44,
|
| 1897 |
+
"learning_rate": 1.605902777777778e-05,
|
| 1898 |
+
"loss": 0.1114,
|
| 1899 |
+
"step": 2730
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 28.54,
|
| 1903 |
+
"learning_rate": 1.5914351851851854e-05,
|
| 1904 |
+
"loss": 0.1157,
|
| 1905 |
+
"step": 2740
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 28.65,
|
| 1909 |
+
"learning_rate": 1.576967592592593e-05,
|
| 1910 |
+
"loss": 0.1322,
|
| 1911 |
+
"step": 2750
|
| 1912 |
+
},
|
| 1913 |
+
{
|
| 1914 |
+
"epoch": 28.75,
|
| 1915 |
+
"learning_rate": 1.5625e-05,
|
| 1916 |
+
"loss": 0.0794,
|
| 1917 |
+
"step": 2760
|
| 1918 |
+
},
|
| 1919 |
+
{
|
| 1920 |
+
"epoch": 28.85,
|
| 1921 |
+
"learning_rate": 1.5480324074074072e-05,
|
| 1922 |
+
"loss": 0.1354,
|
| 1923 |
+
"step": 2770
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 28.96,
|
| 1927 |
+
"learning_rate": 1.533564814814815e-05,
|
| 1928 |
+
"loss": 0.1238,
|
| 1929 |
+
"step": 2780
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 29.0,
|
| 1933 |
+
"eval_accuracy": 0.6018181818181818,
|
| 1934 |
+
"eval_loss": 2.816072702407837,
|
| 1935 |
+
"eval_runtime": 3.8975,
|
| 1936 |
+
"eval_samples_per_second": 141.115,
|
| 1937 |
+
"eval_steps_per_second": 8.98,
|
| 1938 |
+
"step": 2784
|
| 1939 |
+
},
|
| 1940 |
+
{
|
| 1941 |
+
"epoch": 29.06,
|
| 1942 |
+
"learning_rate": 1.5190972222222222e-05,
|
| 1943 |
+
"loss": 0.1339,
|
| 1944 |
+
"step": 2790
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 29.17,
|
| 1948 |
+
"learning_rate": 1.5046296296296297e-05,
|
| 1949 |
+
"loss": 0.098,
|
| 1950 |
+
"step": 2800
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 29.27,
|
| 1954 |
+
"learning_rate": 1.490162037037037e-05,
|
| 1955 |
+
"loss": 0.1411,
|
| 1956 |
+
"step": 2810
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 29.38,
|
| 1960 |
+
"learning_rate": 1.4756944444444445e-05,
|
| 1961 |
+
"loss": 0.0686,
|
| 1962 |
+
"step": 2820
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 29.48,
|
| 1966 |
+
"learning_rate": 1.4612268518518518e-05,
|
| 1967 |
+
"loss": 0.1029,
|
| 1968 |
+
"step": 2830
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 29.58,
|
| 1972 |
+
"learning_rate": 1.4467592592592593e-05,
|
| 1973 |
+
"loss": 0.1361,
|
| 1974 |
+
"step": 2840
|
| 1975 |
+
},
|
| 1976 |
+
{
|
| 1977 |
+
"epoch": 29.69,
|
| 1978 |
+
"learning_rate": 1.4322916666666666e-05,
|
| 1979 |
+
"loss": 0.0927,
|
| 1980 |
+
"step": 2850
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 29.79,
|
| 1984 |
+
"learning_rate": 1.4178240740740741e-05,
|
| 1985 |
+
"loss": 0.1329,
|
| 1986 |
+
"step": 2860
|
| 1987 |
+
},
|
| 1988 |
+
{
|
| 1989 |
+
"epoch": 29.9,
|
| 1990 |
+
"learning_rate": 1.4033564814814815e-05,
|
| 1991 |
+
"loss": 0.1146,
|
| 1992 |
+
"step": 2870
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 30.0,
|
| 1996 |
+
"learning_rate": 1.388888888888889e-05,
|
| 1997 |
+
"loss": 0.1256,
|
| 1998 |
+
"step": 2880
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 30.0,
|
| 2002 |
+
"eval_accuracy": 0.5490909090909091,
|
| 2003 |
+
"eval_loss": 3.491748332977295,
|
| 2004 |
+
"eval_runtime": 3.8988,
|
| 2005 |
+
"eval_samples_per_second": 141.071,
|
| 2006 |
+
"eval_steps_per_second": 8.977,
|
| 2007 |
+
"step": 2880
|
| 2008 |
+
},
|
| 2009 |
+
{
|
| 2010 |
+
"epoch": 30.1,
|
| 2011 |
+
"learning_rate": 1.3744212962962963e-05,
|
| 2012 |
+
"loss": 0.1046,
|
| 2013 |
+
"step": 2890
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 30.21,
|
| 2017 |
+
"learning_rate": 1.3599537037037038e-05,
|
| 2018 |
+
"loss": 0.1199,
|
| 2019 |
+
"step": 2900
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 30.31,
|
| 2023 |
+
"learning_rate": 1.3454861111111111e-05,
|
| 2024 |
+
"loss": 0.1524,
|
| 2025 |
+
"step": 2910
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 30.42,
|
| 2029 |
+
"learning_rate": 1.3310185185185186e-05,
|
| 2030 |
+
"loss": 0.1064,
|
| 2031 |
+
"step": 2920
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 30.52,
|
| 2035 |
+
"learning_rate": 1.316550925925926e-05,
|
| 2036 |
+
"loss": 0.1227,
|
| 2037 |
+
"step": 2930
|
| 2038 |
+
},
|
| 2039 |
+
{
|
| 2040 |
+
"epoch": 30.62,
|
| 2041 |
+
"learning_rate": 1.3020833333333334e-05,
|
| 2042 |
+
"loss": 0.1043,
|
| 2043 |
+
"step": 2940
|
| 2044 |
+
},
|
| 2045 |
+
{
|
| 2046 |
+
"epoch": 30.73,
|
| 2047 |
+
"learning_rate": 1.2876157407407407e-05,
|
| 2048 |
+
"loss": 0.1002,
|
| 2049 |
+
"step": 2950
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 30.83,
|
| 2053 |
+
"learning_rate": 1.2731481481481482e-05,
|
| 2054 |
+
"loss": 0.1153,
|
| 2055 |
+
"step": 2960
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 30.94,
|
| 2059 |
+
"learning_rate": 1.2586805555555556e-05,
|
| 2060 |
+
"loss": 0.1064,
|
| 2061 |
+
"step": 2970
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 31.0,
|
| 2065 |
+
"eval_accuracy": 0.58,
|
| 2066 |
+
"eval_loss": 3.065882682800293,
|
| 2067 |
+
"eval_runtime": 3.8866,
|
| 2068 |
+
"eval_samples_per_second": 141.514,
|
| 2069 |
+
"eval_steps_per_second": 9.005,
|
| 2070 |
+
"step": 2976
|
| 2071 |
+
},
|
| 2072 |
+
{
|
| 2073 |
+
"epoch": 31.04,
|
| 2074 |
+
"learning_rate": 1.244212962962963e-05,
|
| 2075 |
+
"loss": 0.1036,
|
| 2076 |
+
"step": 2980
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 31.15,
|
| 2080 |
+
"learning_rate": 1.2297453703703704e-05,
|
| 2081 |
+
"loss": 0.1149,
|
| 2082 |
+
"step": 2990
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 31.25,
|
| 2086 |
+
"learning_rate": 1.2152777777777779e-05,
|
| 2087 |
+
"loss": 0.1135,
|
| 2088 |
+
"step": 3000
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 31.35,
|
| 2092 |
+
"learning_rate": 1.2008101851851852e-05,
|
| 2093 |
+
"loss": 0.0999,
|
| 2094 |
+
"step": 3010
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"epoch": 31.46,
|
| 2098 |
+
"learning_rate": 1.1863425925925927e-05,
|
| 2099 |
+
"loss": 0.1174,
|
| 2100 |
+
"step": 3020
|
| 2101 |
+
},
|
| 2102 |
+
{
|
| 2103 |
+
"epoch": 31.56,
|
| 2104 |
+
"learning_rate": 1.171875e-05,
|
| 2105 |
+
"loss": 0.1161,
|
| 2106 |
+
"step": 3030
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"epoch": 31.67,
|
| 2110 |
+
"learning_rate": 1.1574074074074075e-05,
|
| 2111 |
+
"loss": 0.1123,
|
| 2112 |
+
"step": 3040
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 31.77,
|
| 2116 |
+
"learning_rate": 1.1429398148148148e-05,
|
| 2117 |
+
"loss": 0.1167,
|
| 2118 |
+
"step": 3050
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 31.88,
|
| 2122 |
+
"learning_rate": 1.1284722222222223e-05,
|
| 2123 |
+
"loss": 0.0939,
|
| 2124 |
+
"step": 3060
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 31.98,
|
| 2128 |
+
"learning_rate": 1.1140046296296297e-05,
|
| 2129 |
+
"loss": 0.1333,
|
| 2130 |
+
"step": 3070
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 32.0,
|
| 2134 |
+
"eval_accuracy": 0.5472727272727272,
|
| 2135 |
+
"eval_loss": 3.5971763134002686,
|
| 2136 |
+
"eval_runtime": 3.9815,
|
| 2137 |
+
"eval_samples_per_second": 138.137,
|
| 2138 |
+
"eval_steps_per_second": 8.791,
|
| 2139 |
+
"step": 3072
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 32.08,
|
| 2143 |
+
"learning_rate": 1.0995370370370372e-05,
|
| 2144 |
+
"loss": 0.1048,
|
| 2145 |
+
"step": 3080
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 32.19,
|
| 2149 |
+
"learning_rate": 1.0850694444444445e-05,
|
| 2150 |
+
"loss": 0.118,
|
| 2151 |
+
"step": 3090
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 32.29,
|
| 2155 |
+
"learning_rate": 1.070601851851852e-05,
|
| 2156 |
+
"loss": 0.1366,
|
| 2157 |
+
"step": 3100
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 32.4,
|
| 2161 |
+
"learning_rate": 1.0561342592592593e-05,
|
| 2162 |
+
"loss": 0.0787,
|
| 2163 |
+
"step": 3110
|
| 2164 |
+
},
|
| 2165 |
+
{
|
| 2166 |
+
"epoch": 32.5,
|
| 2167 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 2168 |
+
"loss": 0.1368,
|
| 2169 |
+
"step": 3120
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 32.6,
|
| 2173 |
+
"learning_rate": 1.0271990740740741e-05,
|
| 2174 |
+
"loss": 0.1137,
|
| 2175 |
+
"step": 3130
|
| 2176 |
+
},
|
| 2177 |
+
{
|
| 2178 |
+
"epoch": 32.71,
|
| 2179 |
+
"learning_rate": 1.0127314814814816e-05,
|
| 2180 |
+
"loss": 0.1161,
|
| 2181 |
+
"step": 3140
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 32.81,
|
| 2185 |
+
"learning_rate": 9.98263888888889e-06,
|
| 2186 |
+
"loss": 0.1096,
|
| 2187 |
+
"step": 3150
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 32.92,
|
| 2191 |
+
"learning_rate": 9.837962962962964e-06,
|
| 2192 |
+
"loss": 0.1134,
|
| 2193 |
+
"step": 3160
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 33.0,
|
| 2197 |
+
"eval_accuracy": 0.54,
|
| 2198 |
+
"eval_loss": 3.611628293991089,
|
| 2199 |
+
"eval_runtime": 3.8603,
|
| 2200 |
+
"eval_samples_per_second": 142.477,
|
| 2201 |
+
"eval_steps_per_second": 9.067,
|
| 2202 |
+
"step": 3168
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 33.02,
|
| 2206 |
+
"learning_rate": 9.693287037037038e-06,
|
| 2207 |
+
"loss": 0.0905,
|
| 2208 |
+
"step": 3170
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 33.12,
|
| 2212 |
+
"learning_rate": 9.548611111111111e-06,
|
| 2213 |
+
"loss": 0.1007,
|
| 2214 |
+
"step": 3180
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 33.23,
|
| 2218 |
+
"learning_rate": 9.403935185185186e-06,
|
| 2219 |
+
"loss": 0.1342,
|
| 2220 |
+
"step": 3190
|
| 2221 |
+
},
|
| 2222 |
+
{
|
| 2223 |
+
"epoch": 33.33,
|
| 2224 |
+
"learning_rate": 9.259259259259259e-06,
|
| 2225 |
+
"loss": 0.1072,
|
| 2226 |
+
"step": 3200
|
| 2227 |
+
},
|
| 2228 |
+
{
|
| 2229 |
+
"epoch": 33.44,
|
| 2230 |
+
"learning_rate": 9.114583333333334e-06,
|
| 2231 |
+
"loss": 0.0824,
|
| 2232 |
+
"step": 3210
|
| 2233 |
+
},
|
| 2234 |
+
{
|
| 2235 |
+
"epoch": 33.54,
|
| 2236 |
+
"learning_rate": 8.969907407407407e-06,
|
| 2237 |
+
"loss": 0.123,
|
| 2238 |
+
"step": 3220
|
| 2239 |
+
},
|
| 2240 |
+
{
|
| 2241 |
+
"epoch": 33.65,
|
| 2242 |
+
"learning_rate": 8.825231481481482e-06,
|
| 2243 |
+
"loss": 0.1083,
|
| 2244 |
+
"step": 3230
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 33.75,
|
| 2248 |
+
"learning_rate": 8.680555555555556e-06,
|
| 2249 |
+
"loss": 0.1073,
|
| 2250 |
+
"step": 3240
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 33.85,
|
| 2254 |
+
"learning_rate": 8.53587962962963e-06,
|
| 2255 |
+
"loss": 0.0854,
|
| 2256 |
+
"step": 3250
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 33.96,
|
| 2260 |
+
"learning_rate": 8.391203703703704e-06,
|
| 2261 |
+
"loss": 0.0831,
|
| 2262 |
+
"step": 3260
|
| 2263 |
+
},
|
| 2264 |
+
{
|
| 2265 |
+
"epoch": 34.0,
|
| 2266 |
+
"eval_accuracy": 0.5509090909090909,
|
| 2267 |
+
"eval_loss": 3.530831813812256,
|
| 2268 |
+
"eval_runtime": 3.9226,
|
| 2269 |
+
"eval_samples_per_second": 140.214,
|
| 2270 |
+
"eval_steps_per_second": 8.923,
|
| 2271 |
+
"step": 3264
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 34.06,
|
| 2275 |
+
"learning_rate": 8.246527777777779e-06,
|
| 2276 |
+
"loss": 0.1373,
|
| 2277 |
+
"step": 3270
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 34.17,
|
| 2281 |
+
"learning_rate": 8.101851851851852e-06,
|
| 2282 |
+
"loss": 0.0997,
|
| 2283 |
+
"step": 3280
|
| 2284 |
+
},
|
| 2285 |
+
{
|
| 2286 |
+
"epoch": 34.27,
|
| 2287 |
+
"learning_rate": 7.957175925925927e-06,
|
| 2288 |
+
"loss": 0.0959,
|
| 2289 |
+
"step": 3290
|
| 2290 |
+
},
|
| 2291 |
+
{
|
| 2292 |
+
"epoch": 34.38,
|
| 2293 |
+
"learning_rate": 7.8125e-06,
|
| 2294 |
+
"loss": 0.0999,
|
| 2295 |
+
"step": 3300
|
| 2296 |
+
},
|
| 2297 |
+
{
|
| 2298 |
+
"epoch": 34.48,
|
| 2299 |
+
"learning_rate": 7.667824074074075e-06,
|
| 2300 |
+
"loss": 0.0585,
|
| 2301 |
+
"step": 3310
|
| 2302 |
+
},
|
| 2303 |
+
{
|
| 2304 |
+
"epoch": 34.58,
|
| 2305 |
+
"learning_rate": 7.523148148148148e-06,
|
| 2306 |
+
"loss": 0.0815,
|
| 2307 |
+
"step": 3320
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 34.69,
|
| 2311 |
+
"learning_rate": 7.3784722222222225e-06,
|
| 2312 |
+
"loss": 0.073,
|
| 2313 |
+
"step": 3330
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 34.79,
|
| 2317 |
+
"learning_rate": 7.2337962962962966e-06,
|
| 2318 |
+
"loss": 0.1152,
|
| 2319 |
+
"step": 3340
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 34.9,
|
| 2323 |
+
"learning_rate": 7.089120370370371e-06,
|
| 2324 |
+
"loss": 0.0926,
|
| 2325 |
+
"step": 3350
|
| 2326 |
+
},
|
| 2327 |
+
{
|
| 2328 |
+
"epoch": 35.0,
|
| 2329 |
+
"learning_rate": 6.944444444444445e-06,
|
| 2330 |
+
"loss": 0.1035,
|
| 2331 |
+
"step": 3360
|
| 2332 |
+
},
|
| 2333 |
+
{
|
| 2334 |
+
"epoch": 35.0,
|
| 2335 |
+
"eval_accuracy": 0.5581818181818182,
|
| 2336 |
+
"eval_loss": 3.4789416790008545,
|
| 2337 |
+
"eval_runtime": 3.9255,
|
| 2338 |
+
"eval_samples_per_second": 140.108,
|
| 2339 |
+
"eval_steps_per_second": 8.916,
|
| 2340 |
+
"step": 3360
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 35.1,
|
| 2344 |
+
"learning_rate": 6.799768518518519e-06,
|
| 2345 |
+
"loss": 0.1094,
|
| 2346 |
+
"step": 3370
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"epoch": 35.21,
|
| 2350 |
+
"learning_rate": 6.655092592592593e-06,
|
| 2351 |
+
"loss": 0.1014,
|
| 2352 |
+
"step": 3380
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 35.31,
|
| 2356 |
+
"learning_rate": 6.510416666666667e-06,
|
| 2357 |
+
"loss": 0.0975,
|
| 2358 |
+
"step": 3390
|
| 2359 |
+
},
|
| 2360 |
+
{
|
| 2361 |
+
"epoch": 35.42,
|
| 2362 |
+
"learning_rate": 6.365740740740741e-06,
|
| 2363 |
+
"loss": 0.0769,
|
| 2364 |
+
"step": 3400
|
| 2365 |
+
},
|
| 2366 |
+
{
|
| 2367 |
+
"epoch": 35.52,
|
| 2368 |
+
"learning_rate": 6.221064814814815e-06,
|
| 2369 |
+
"loss": 0.0742,
|
| 2370 |
+
"step": 3410
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 35.62,
|
| 2374 |
+
"learning_rate": 6.076388888888889e-06,
|
| 2375 |
+
"loss": 0.0994,
|
| 2376 |
+
"step": 3420
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 35.73,
|
| 2380 |
+
"learning_rate": 5.9317129629629635e-06,
|
| 2381 |
+
"loss": 0.0755,
|
| 2382 |
+
"step": 3430
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 35.83,
|
| 2386 |
+
"learning_rate": 5.787037037037038e-06,
|
| 2387 |
+
"loss": 0.0833,
|
| 2388 |
+
"step": 3440
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"epoch": 35.94,
|
| 2392 |
+
"learning_rate": 5.642361111111112e-06,
|
| 2393 |
+
"loss": 0.0957,
|
| 2394 |
+
"step": 3450
|
| 2395 |
+
},
|
| 2396 |
+
{
|
| 2397 |
+
"epoch": 36.0,
|
| 2398 |
+
"eval_accuracy": 0.5509090909090909,
|
| 2399 |
+
"eval_loss": 3.6358087062835693,
|
| 2400 |
+
"eval_runtime": 3.9255,
|
| 2401 |
+
"eval_samples_per_second": 140.108,
|
| 2402 |
+
"eval_steps_per_second": 8.916,
|
| 2403 |
+
"step": 3456
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 36.04,
|
| 2407 |
+
"learning_rate": 5.497685185185186e-06,
|
| 2408 |
+
"loss": 0.1083,
|
| 2409 |
+
"step": 3460
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 36.15,
|
| 2413 |
+
"learning_rate": 5.35300925925926e-06,
|
| 2414 |
+
"loss": 0.0927,
|
| 2415 |
+
"step": 3470
|
| 2416 |
+
},
|
| 2417 |
+
{
|
| 2418 |
+
"epoch": 36.25,
|
| 2419 |
+
"learning_rate": 5.208333333333334e-06,
|
| 2420 |
+
"loss": 0.0876,
|
| 2421 |
+
"step": 3480
|
| 2422 |
+
},
|
| 2423 |
+
{
|
| 2424 |
+
"epoch": 36.35,
|
| 2425 |
+
"learning_rate": 5.063657407407408e-06,
|
| 2426 |
+
"loss": 0.0921,
|
| 2427 |
+
"step": 3490
|
| 2428 |
+
},
|
| 2429 |
+
{
|
| 2430 |
+
"epoch": 36.46,
|
| 2431 |
+
"learning_rate": 4.918981481481482e-06,
|
| 2432 |
+
"loss": 0.0909,
|
| 2433 |
+
"step": 3500
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 36.56,
|
| 2437 |
+
"learning_rate": 4.7743055555555555e-06,
|
| 2438 |
+
"loss": 0.0927,
|
| 2439 |
+
"step": 3510
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 36.67,
|
| 2443 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 2444 |
+
"loss": 0.1031,
|
| 2445 |
+
"step": 3520
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 36.77,
|
| 2449 |
+
"learning_rate": 4.484953703703704e-06,
|
| 2450 |
+
"loss": 0.1017,
|
| 2451 |
+
"step": 3530
|
| 2452 |
+
},
|
| 2453 |
+
{
|
| 2454 |
+
"epoch": 36.88,
|
| 2455 |
+
"learning_rate": 4.340277777777778e-06,
|
| 2456 |
+
"loss": 0.064,
|
| 2457 |
+
"step": 3540
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 36.98,
|
| 2461 |
+
"learning_rate": 4.195601851851852e-06,
|
| 2462 |
+
"loss": 0.0764,
|
| 2463 |
+
"step": 3550
|
| 2464 |
+
},
|
| 2465 |
+
{
|
| 2466 |
+
"epoch": 37.0,
|
| 2467 |
+
"eval_accuracy": 0.5709090909090909,
|
| 2468 |
+
"eval_loss": 3.363851547241211,
|
| 2469 |
+
"eval_runtime": 3.9482,
|
| 2470 |
+
"eval_samples_per_second": 139.304,
|
| 2471 |
+
"eval_steps_per_second": 8.865,
|
| 2472 |
+
"step": 3552
|
| 2473 |
+
},
|
| 2474 |
+
{
|
| 2475 |
+
"epoch": 37.08,
|
| 2476 |
+
"learning_rate": 4.050925925925926e-06,
|
| 2477 |
+
"loss": 0.1104,
|
| 2478 |
+
"step": 3560
|
| 2479 |
+
},
|
| 2480 |
+
{
|
| 2481 |
+
"epoch": 37.19,
|
| 2482 |
+
"learning_rate": 3.90625e-06,
|
| 2483 |
+
"loss": 0.0785,
|
| 2484 |
+
"step": 3570
|
| 2485 |
+
},
|
| 2486 |
+
{
|
| 2487 |
+
"epoch": 37.29,
|
| 2488 |
+
"learning_rate": 3.761574074074074e-06,
|
| 2489 |
+
"loss": 0.0777,
|
| 2490 |
+
"step": 3580
|
| 2491 |
+
},
|
| 2492 |
+
{
|
| 2493 |
+
"epoch": 37.4,
|
| 2494 |
+
"learning_rate": 3.6168981481481483e-06,
|
| 2495 |
+
"loss": 0.0696,
|
| 2496 |
+
"step": 3590
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 37.5,
|
| 2500 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 2501 |
+
"loss": 0.0789,
|
| 2502 |
+
"step": 3600
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 37.6,
|
| 2506 |
+
"learning_rate": 3.3275462962962965e-06,
|
| 2507 |
+
"loss": 0.0921,
|
| 2508 |
+
"step": 3610
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 37.71,
|
| 2512 |
+
"learning_rate": 3.1828703703703706e-06,
|
| 2513 |
+
"loss": 0.0932,
|
| 2514 |
+
"step": 3620
|
| 2515 |
+
},
|
| 2516 |
+
{
|
| 2517 |
+
"epoch": 37.81,
|
| 2518 |
+
"learning_rate": 3.0381944444444447e-06,
|
| 2519 |
+
"loss": 0.0614,
|
| 2520 |
+
"step": 3630
|
| 2521 |
+
},
|
| 2522 |
+
{
|
| 2523 |
+
"epoch": 37.92,
|
| 2524 |
+
"learning_rate": 2.893518518518519e-06,
|
| 2525 |
+
"loss": 0.072,
|
| 2526 |
+
"step": 3640
|
| 2527 |
+
},
|
| 2528 |
+
{
|
| 2529 |
+
"epoch": 38.0,
|
| 2530 |
+
"eval_accuracy": 0.5563636363636364,
|
| 2531 |
+
"eval_loss": 3.5639150142669678,
|
| 2532 |
+
"eval_runtime": 3.9387,
|
| 2533 |
+
"eval_samples_per_second": 139.642,
|
| 2534 |
+
"eval_steps_per_second": 8.886,
|
| 2535 |
+
"step": 3648
|
| 2536 |
+
},
|
| 2537 |
+
{
|
| 2538 |
+
"epoch": 38.02,
|
| 2539 |
+
"learning_rate": 2.748842592592593e-06,
|
| 2540 |
+
"loss": 0.1215,
|
| 2541 |
+
"step": 3650
|
| 2542 |
+
},
|
| 2543 |
+
{
|
| 2544 |
+
"epoch": 38.12,
|
| 2545 |
+
"learning_rate": 2.604166666666667e-06,
|
| 2546 |
+
"loss": 0.0942,
|
| 2547 |
+
"step": 3660
|
| 2548 |
+
},
|
| 2549 |
+
{
|
| 2550 |
+
"epoch": 38.23,
|
| 2551 |
+
"learning_rate": 2.459490740740741e-06,
|
| 2552 |
+
"loss": 0.0525,
|
| 2553 |
+
"step": 3670
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 38.33,
|
| 2557 |
+
"learning_rate": 2.3148148148148148e-06,
|
| 2558 |
+
"loss": 0.0871,
|
| 2559 |
+
"step": 3680
|
| 2560 |
+
},
|
| 2561 |
+
{
|
| 2562 |
+
"epoch": 38.44,
|
| 2563 |
+
"learning_rate": 2.170138888888889e-06,
|
| 2564 |
+
"loss": 0.0973,
|
| 2565 |
+
"step": 3690
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 38.54,
|
| 2569 |
+
"learning_rate": 2.025462962962963e-06,
|
| 2570 |
+
"loss": 0.1067,
|
| 2571 |
+
"step": 3700
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 38.65,
|
| 2575 |
+
"learning_rate": 1.880787037037037e-06,
|
| 2576 |
+
"loss": 0.1219,
|
| 2577 |
+
"step": 3710
|
| 2578 |
+
},
|
| 2579 |
+
{
|
| 2580 |
+
"epoch": 38.75,
|
| 2581 |
+
"learning_rate": 1.7361111111111112e-06,
|
| 2582 |
+
"loss": 0.0806,
|
| 2583 |
+
"step": 3720
|
| 2584 |
+
},
|
| 2585 |
+
{
|
| 2586 |
+
"epoch": 38.85,
|
| 2587 |
+
"learning_rate": 1.5914351851851853e-06,
|
| 2588 |
+
"loss": 0.0646,
|
| 2589 |
+
"step": 3730
|
| 2590 |
+
},
|
| 2591 |
+
{
|
| 2592 |
+
"epoch": 38.96,
|
| 2593 |
+
"learning_rate": 1.4467592592592594e-06,
|
| 2594 |
+
"loss": 0.0727,
|
| 2595 |
+
"step": 3740
|
| 2596 |
+
},
|
| 2597 |
+
{
|
| 2598 |
+
"epoch": 39.0,
|
| 2599 |
+
"eval_accuracy": 0.5581818181818182,
|
| 2600 |
+
"eval_loss": 3.519263744354248,
|
| 2601 |
+
"eval_runtime": 3.9518,
|
| 2602 |
+
"eval_samples_per_second": 139.177,
|
| 2603 |
+
"eval_steps_per_second": 8.857,
|
| 2604 |
+
"step": 3744
|
| 2605 |
+
},
|
| 2606 |
+
{
|
| 2607 |
+
"epoch": 39.06,
|
| 2608 |
+
"learning_rate": 1.3020833333333335e-06,
|
| 2609 |
+
"loss": 0.1025,
|
| 2610 |
+
"step": 3750
|
| 2611 |
+
},
|
| 2612 |
+
{
|
| 2613 |
+
"epoch": 39.17,
|
| 2614 |
+
"learning_rate": 1.1574074074074074e-06,
|
| 2615 |
+
"loss": 0.1064,
|
| 2616 |
+
"step": 3760
|
| 2617 |
+
},
|
| 2618 |
+
{
|
| 2619 |
+
"epoch": 39.27,
|
| 2620 |
+
"learning_rate": 1.0127314814814815e-06,
|
| 2621 |
+
"loss": 0.0921,
|
| 2622 |
+
"step": 3770
|
| 2623 |
+
},
|
| 2624 |
+
{
|
| 2625 |
+
"epoch": 39.38,
|
| 2626 |
+
"learning_rate": 8.680555555555556e-07,
|
| 2627 |
+
"loss": 0.0786,
|
| 2628 |
+
"step": 3780
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 39.48,
|
| 2632 |
+
"learning_rate": 7.233796296296297e-07,
|
| 2633 |
+
"loss": 0.0767,
|
| 2634 |
+
"step": 3790
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 39.58,
|
| 2638 |
+
"learning_rate": 5.787037037037037e-07,
|
| 2639 |
+
"loss": 0.0661,
|
| 2640 |
+
"step": 3800
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 39.69,
|
| 2644 |
+
"learning_rate": 4.340277777777778e-07,
|
| 2645 |
+
"loss": 0.0873,
|
| 2646 |
+
"step": 3810
|
| 2647 |
+
},
|
| 2648 |
+
{
|
| 2649 |
+
"epoch": 39.79,
|
| 2650 |
+
"learning_rate": 2.8935185185185185e-07,
|
| 2651 |
+
"loss": 0.0692,
|
| 2652 |
+
"step": 3820
|
| 2653 |
+
},
|
| 2654 |
+
{
|
| 2655 |
+
"epoch": 39.9,
|
| 2656 |
+
"learning_rate": 1.4467592592592592e-07,
|
| 2657 |
+
"loss": 0.1063,
|
| 2658 |
+
"step": 3830
|
| 2659 |
+
},
|
| 2660 |
+
{
|
| 2661 |
+
"epoch": 40.0,
|
| 2662 |
+
"learning_rate": 0.0,
|
| 2663 |
+
"loss": 0.0619,
|
| 2664 |
+
"step": 3840
|
| 2665 |
+
},
|
| 2666 |
+
{
|
| 2667 |
+
"epoch": 40.0,
|
| 2668 |
+
"eval_accuracy": 0.5581818181818182,
|
| 2669 |
+
"eval_loss": 3.5836477279663086,
|
| 2670 |
+
"eval_runtime": 3.9696,
|
| 2671 |
+
"eval_samples_per_second": 138.553,
|
| 2672 |
+
"eval_steps_per_second": 8.817,
|
| 2673 |
+
"step": 3840
|
| 2674 |
+
},
|
| 2675 |
+
{
|
| 2676 |
+
"epoch": 40.0,
|
| 2677 |
+
"step": 3840,
|
| 2678 |
+
"total_flos": 1.9002720805055447e+19,
|
| 2679 |
+
"train_loss": 0.31003439707371094,
|
| 2680 |
+
"train_runtime": 4246.6951,
|
| 2681 |
+
"train_samples_per_second": 57.767,
|
| 2682 |
+
"train_steps_per_second": 0.904
|
| 2683 |
+
}
|
| 2684 |
+
],
|
| 2685 |
+
"logging_steps": 10,
|
| 2686 |
+
"max_steps": 3840,
|
| 2687 |
+
"num_input_tokens_seen": 0,
|
| 2688 |
+
"num_train_epochs": 40,
|
| 2689 |
+
"save_steps": 500,
|
| 2690 |
+
"total_flos": 1.9002720805055447e+19,
|
| 2691 |
+
"train_batch_size": 16,
|
| 2692 |
+
"trial_name": null,
|
| 2693 |
+
"trial_params": null
|
| 2694 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e844911265167931cf7febb1bde2ea95ae6c121505c9800dd15692d7dc4867f
|
| 3 |
+
size 4728
|