End of training
Browse files- README.md +70 -0
- all_results.json +13 -0
- config.json +38 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- runs/Aug01_11-36-26_d92e2767e41c/events.out.tfevents.1722512195.d92e2767e41c.1213.0 +3 -0
- runs/Aug01_11-36-26_d92e2767e41c/events.out.tfevents.1722512867.d92e2767e41c.1213.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +832 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: google/vit-base-patch16-224-in21k
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
metrics:
|
7 |
+
- accuracy
|
8 |
+
model-index:
|
9 |
+
- name: vit-base-images
|
10 |
+
results: []
|
11 |
+
---
|
12 |
+
|
13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
+
should probably proofread and complete it, then remove this comment. -->
|
15 |
+
|
16 |
+
# vit-base-images
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 0.3465
|
21 |
+
- Accuracy: 0.905
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training procedure
|
36 |
+
|
37 |
+
### Training hyperparameters
|
38 |
+
|
39 |
+
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 0.0002
|
41 |
+
- train_batch_size: 16
|
42 |
+
- eval_batch_size: 8
|
43 |
+
- seed: 42
|
44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
+
- lr_scheduler_type: linear
|
46 |
+
- num_epochs: 4
|
47 |
+
- mixed_precision_training: Native AMP
|
48 |
+
|
49 |
+
### Training results
|
50 |
+
|
51 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
52 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
53 |
+
| 0.7334 | 0.4 | 100 | 0.6142 | 0.779 |
|
54 |
+
| 0.6032 | 0.8 | 200 | 0.5516 | 0.808 |
|
55 |
+
| 0.4725 | 1.2 | 300 | 0.4390 | 0.854 |
|
56 |
+
| 0.3638 | 1.6 | 400 | 0.4622 | 0.822 |
|
57 |
+
| 0.3279 | 2.0 | 500 | 0.3772 | 0.876 |
|
58 |
+
| 0.1337 | 2.4 | 600 | 0.4518 | 0.869 |
|
59 |
+
| 0.236 | 2.8 | 700 | 0.3766 | 0.878 |
|
60 |
+
| 0.0275 | 3.2 | 800 | 0.3518 | 0.891 |
|
61 |
+
| 0.0427 | 3.6 | 900 | 0.3709 | 0.896 |
|
62 |
+
| 0.0363 | 4.0 | 1000 | 0.3465 | 0.905 |
|
63 |
+
|
64 |
+
|
65 |
+
### Framework versions
|
66 |
+
|
67 |
+
- Transformers 4.42.4
|
68 |
+
- Pytorch 2.3.1+cu121
|
69 |
+
- Datasets 2.20.0
|
70 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.905,
|
4 |
+
"eval_loss": 0.34654033184051514,
|
5 |
+
"eval_runtime": 21.2431,
|
6 |
+
"eval_samples_per_second": 47.074,
|
7 |
+
"eval_steps_per_second": 5.884,
|
8 |
+
"total_flos": 1.239905171570688e+18,
|
9 |
+
"train_loss": 0.3179253642559052,
|
10 |
+
"train_runtime": 607.1163,
|
11 |
+
"train_samples_per_second": 26.354,
|
12 |
+
"train_steps_per_second": 1.647
|
13 |
+
}
|
config.json
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/vit-base-patch16-224-in21k",
|
3 |
+
"architectures": [
|
4 |
+
"ViTForImageClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.0,
|
7 |
+
"encoder_stride": 16,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_dropout_prob": 0.0,
|
10 |
+
"hidden_size": 768,
|
11 |
+
"id2label": {
|
12 |
+
"0": "basal_cell_carcinoma",
|
13 |
+
"1": "benign_keratosis-like_lesions",
|
14 |
+
"2": "melanocytic_Nevi",
|
15 |
+
"3": "dermatofibroma",
|
16 |
+
"4": "actinic_keratoses"
|
17 |
+
},
|
18 |
+
"image_size": 224,
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 3072,
|
21 |
+
"label2id": {
|
22 |
+
"actinic_keratoses": "4",
|
23 |
+
"basal_cell_carcinoma": "0",
|
24 |
+
"benign_keratosis-like_lesions": "1",
|
25 |
+
"dermatofibroma": "3",
|
26 |
+
"melanocytic_Nevi": "2"
|
27 |
+
},
|
28 |
+
"layer_norm_eps": 1e-12,
|
29 |
+
"model_type": "vit",
|
30 |
+
"num_attention_heads": 12,
|
31 |
+
"num_channels": 3,
|
32 |
+
"num_hidden_layers": 12,
|
33 |
+
"patch_size": 16,
|
34 |
+
"problem_type": "single_label_classification",
|
35 |
+
"qkv_bias": true,
|
36 |
+
"torch_dtype": "float32",
|
37 |
+
"transformers_version": "4.42.4"
|
38 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.905,
|
4 |
+
"eval_loss": 0.34654033184051514,
|
5 |
+
"eval_runtime": 21.2431,
|
6 |
+
"eval_samples_per_second": 47.074,
|
7 |
+
"eval_steps_per_second": 5.884
|
8 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:65bf56253dc41fb2232cd24bf6b31aabbe82f9a83128a3285ee7d833751a62a9
|
3 |
+
size 343233204
|
preprocessor_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"do_rescale": true,
|
4 |
+
"do_resize": true,
|
5 |
+
"image_mean": [
|
6 |
+
0.5,
|
7 |
+
0.5,
|
8 |
+
0.5
|
9 |
+
],
|
10 |
+
"image_processor_type": "ViTImageProcessor",
|
11 |
+
"image_std": [
|
12 |
+
0.5,
|
13 |
+
0.5,
|
14 |
+
0.5
|
15 |
+
],
|
16 |
+
"resample": 2,
|
17 |
+
"rescale_factor": 0.00392156862745098,
|
18 |
+
"size": {
|
19 |
+
"height": 224,
|
20 |
+
"width": 224
|
21 |
+
}
|
22 |
+
}
|
runs/Aug01_11-36-26_d92e2767e41c/events.out.tfevents.1722512195.d92e2767e41c.1213.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2ae74cc946981398b548d8f98529af80bb432abbb2cb17c83bb809adff88c3c9
|
3 |
+
size 29704
|
runs/Aug01_11-36-26_d92e2767e41c/events.out.tfevents.1722512867.d92e2767e41c.1213.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2f5de4c4748747364501c94017e281b28cd21c34f5ae95907087b86abac134d9
|
3 |
+
size 411
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"total_flos": 1.239905171570688e+18,
|
4 |
+
"train_loss": 0.3179253642559052,
|
5 |
+
"train_runtime": 607.1163,
|
6 |
+
"train_samples_per_second": 26.354,
|
7 |
+
"train_steps_per_second": 1.647
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,832 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.34654033184051514,
|
3 |
+
"best_model_checkpoint": "./vit-base-images/checkpoint-1000",
|
4 |
+
"epoch": 4.0,
|
5 |
+
"eval_steps": 100,
|
6 |
+
"global_step": 1000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.04,
|
13 |
+
"grad_norm": 1.5996569395065308,
|
14 |
+
"learning_rate": 0.00019800000000000002,
|
15 |
+
"loss": 1.0932,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.08,
|
20 |
+
"grad_norm": 1.9198634624481201,
|
21 |
+
"learning_rate": 0.000196,
|
22 |
+
"loss": 0.9522,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.12,
|
27 |
+
"grad_norm": 1.2456237077713013,
|
28 |
+
"learning_rate": 0.000194,
|
29 |
+
"loss": 0.6875,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.16,
|
34 |
+
"grad_norm": 1.7687036991119385,
|
35 |
+
"learning_rate": 0.000192,
|
36 |
+
"loss": 0.9009,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.2,
|
41 |
+
"grad_norm": 2.880617141723633,
|
42 |
+
"learning_rate": 0.00019,
|
43 |
+
"loss": 0.7155,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.24,
|
48 |
+
"grad_norm": 1.5424929857254028,
|
49 |
+
"learning_rate": 0.000188,
|
50 |
+
"loss": 0.8144,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.28,
|
55 |
+
"grad_norm": 1.816297173500061,
|
56 |
+
"learning_rate": 0.00018600000000000002,
|
57 |
+
"loss": 0.6641,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.32,
|
62 |
+
"grad_norm": 2.0054569244384766,
|
63 |
+
"learning_rate": 0.00018400000000000003,
|
64 |
+
"loss": 0.5917,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.36,
|
69 |
+
"grad_norm": 2.2372283935546875,
|
70 |
+
"learning_rate": 0.000182,
|
71 |
+
"loss": 0.7852,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.4,
|
76 |
+
"grad_norm": 2.542130708694458,
|
77 |
+
"learning_rate": 0.00018,
|
78 |
+
"loss": 0.7334,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.4,
|
83 |
+
"eval_accuracy": 0.779,
|
84 |
+
"eval_loss": 0.6142178773880005,
|
85 |
+
"eval_runtime": 18.2795,
|
86 |
+
"eval_samples_per_second": 54.706,
|
87 |
+
"eval_steps_per_second": 6.838,
|
88 |
+
"step": 100
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 0.44,
|
92 |
+
"grad_norm": 1.4493112564086914,
|
93 |
+
"learning_rate": 0.00017800000000000002,
|
94 |
+
"loss": 0.5289,
|
95 |
+
"step": 110
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 0.48,
|
99 |
+
"grad_norm": 1.905771017074585,
|
100 |
+
"learning_rate": 0.00017600000000000002,
|
101 |
+
"loss": 0.6191,
|
102 |
+
"step": 120
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 0.52,
|
106 |
+
"grad_norm": 2.2236440181732178,
|
107 |
+
"learning_rate": 0.000174,
|
108 |
+
"loss": 0.5111,
|
109 |
+
"step": 130
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 0.56,
|
113 |
+
"grad_norm": 2.113398551940918,
|
114 |
+
"learning_rate": 0.000172,
|
115 |
+
"loss": 0.6606,
|
116 |
+
"step": 140
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.6,
|
120 |
+
"grad_norm": 2.4624953269958496,
|
121 |
+
"learning_rate": 0.00017,
|
122 |
+
"loss": 0.5002,
|
123 |
+
"step": 150
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 0.64,
|
127 |
+
"grad_norm": 2.324570417404175,
|
128 |
+
"learning_rate": 0.000168,
|
129 |
+
"loss": 0.9353,
|
130 |
+
"step": 160
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 0.68,
|
134 |
+
"grad_norm": 5.384814262390137,
|
135 |
+
"learning_rate": 0.000166,
|
136 |
+
"loss": 0.6604,
|
137 |
+
"step": 170
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.72,
|
141 |
+
"grad_norm": 0.8541224598884583,
|
142 |
+
"learning_rate": 0.000164,
|
143 |
+
"loss": 0.4894,
|
144 |
+
"step": 180
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.76,
|
148 |
+
"grad_norm": 3.017305612564087,
|
149 |
+
"learning_rate": 0.000162,
|
150 |
+
"loss": 0.6219,
|
151 |
+
"step": 190
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 0.8,
|
155 |
+
"grad_norm": 1.9483362436294556,
|
156 |
+
"learning_rate": 0.00016,
|
157 |
+
"loss": 0.6032,
|
158 |
+
"step": 200
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.8,
|
162 |
+
"eval_accuracy": 0.808,
|
163 |
+
"eval_loss": 0.5516341328620911,
|
164 |
+
"eval_runtime": 14.5864,
|
165 |
+
"eval_samples_per_second": 68.557,
|
166 |
+
"eval_steps_per_second": 8.57,
|
167 |
+
"step": 200
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 0.84,
|
171 |
+
"grad_norm": 2.7376227378845215,
|
172 |
+
"learning_rate": 0.00015800000000000002,
|
173 |
+
"loss": 0.4968,
|
174 |
+
"step": 210
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 0.88,
|
178 |
+
"grad_norm": 1.563944697380066,
|
179 |
+
"learning_rate": 0.00015600000000000002,
|
180 |
+
"loss": 0.4505,
|
181 |
+
"step": 220
|
182 |
+
},
|
183 |
+
{
|
184 |
+
"epoch": 0.92,
|
185 |
+
"grad_norm": 1.3606369495391846,
|
186 |
+
"learning_rate": 0.000154,
|
187 |
+
"loss": 0.5368,
|
188 |
+
"step": 230
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.96,
|
192 |
+
"grad_norm": 1.3428421020507812,
|
193 |
+
"learning_rate": 0.000152,
|
194 |
+
"loss": 0.4932,
|
195 |
+
"step": 240
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 1.0,
|
199 |
+
"grad_norm": 1.9562724828720093,
|
200 |
+
"learning_rate": 0.00015000000000000001,
|
201 |
+
"loss": 0.4884,
|
202 |
+
"step": 250
|
203 |
+
},
|
204 |
+
{
|
205 |
+
"epoch": 1.04,
|
206 |
+
"grad_norm": 0.947496771812439,
|
207 |
+
"learning_rate": 0.000148,
|
208 |
+
"loss": 0.381,
|
209 |
+
"step": 260
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 1.08,
|
213 |
+
"grad_norm": 1.6039777994155884,
|
214 |
+
"learning_rate": 0.000146,
|
215 |
+
"loss": 0.6633,
|
216 |
+
"step": 270
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 1.12,
|
220 |
+
"grad_norm": 1.8116464614868164,
|
221 |
+
"learning_rate": 0.000144,
|
222 |
+
"loss": 0.3728,
|
223 |
+
"step": 280
|
224 |
+
},
|
225 |
+
{
|
226 |
+
"epoch": 1.16,
|
227 |
+
"grad_norm": 1.6644967794418335,
|
228 |
+
"learning_rate": 0.000142,
|
229 |
+
"loss": 0.3299,
|
230 |
+
"step": 290
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 1.2,
|
234 |
+
"grad_norm": 1.5359082221984863,
|
235 |
+
"learning_rate": 0.00014,
|
236 |
+
"loss": 0.4725,
|
237 |
+
"step": 300
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 1.2,
|
241 |
+
"eval_accuracy": 0.854,
|
242 |
+
"eval_loss": 0.43897509574890137,
|
243 |
+
"eval_runtime": 14.256,
|
244 |
+
"eval_samples_per_second": 70.146,
|
245 |
+
"eval_steps_per_second": 8.768,
|
246 |
+
"step": 300
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 1.24,
|
250 |
+
"grad_norm": 2.018160581588745,
|
251 |
+
"learning_rate": 0.000138,
|
252 |
+
"loss": 0.3064,
|
253 |
+
"step": 310
|
254 |
+
},
|
255 |
+
{
|
256 |
+
"epoch": 1.28,
|
257 |
+
"grad_norm": 1.5475637912750244,
|
258 |
+
"learning_rate": 0.00013600000000000003,
|
259 |
+
"loss": 0.2928,
|
260 |
+
"step": 320
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 1.32,
|
264 |
+
"grad_norm": 2.780301809310913,
|
265 |
+
"learning_rate": 0.000134,
|
266 |
+
"loss": 0.2959,
|
267 |
+
"step": 330
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 1.3599999999999999,
|
271 |
+
"grad_norm": 1.0915693044662476,
|
272 |
+
"learning_rate": 0.000132,
|
273 |
+
"loss": 0.3152,
|
274 |
+
"step": 340
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 1.4,
|
278 |
+
"grad_norm": 2.1470773220062256,
|
279 |
+
"learning_rate": 0.00013000000000000002,
|
280 |
+
"loss": 0.4123,
|
281 |
+
"step": 350
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 1.44,
|
285 |
+
"grad_norm": 4.054312705993652,
|
286 |
+
"learning_rate": 0.00012800000000000002,
|
287 |
+
"loss": 0.5676,
|
288 |
+
"step": 360
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 1.48,
|
292 |
+
"grad_norm": 1.8798813819885254,
|
293 |
+
"learning_rate": 0.000126,
|
294 |
+
"loss": 0.3909,
|
295 |
+
"step": 370
|
296 |
+
},
|
297 |
+
{
|
298 |
+
"epoch": 1.52,
|
299 |
+
"grad_norm": 2.3789453506469727,
|
300 |
+
"learning_rate": 0.000124,
|
301 |
+
"loss": 0.419,
|
302 |
+
"step": 380
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 1.56,
|
306 |
+
"grad_norm": 1.7660586833953857,
|
307 |
+
"learning_rate": 0.000122,
|
308 |
+
"loss": 0.516,
|
309 |
+
"step": 390
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 1.6,
|
313 |
+
"grad_norm": 3.304502010345459,
|
314 |
+
"learning_rate": 0.00012,
|
315 |
+
"loss": 0.3638,
|
316 |
+
"step": 400
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 1.6,
|
320 |
+
"eval_accuracy": 0.822,
|
321 |
+
"eval_loss": 0.4622470438480377,
|
322 |
+
"eval_runtime": 14.2766,
|
323 |
+
"eval_samples_per_second": 70.045,
|
324 |
+
"eval_steps_per_second": 8.756,
|
325 |
+
"step": 400
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 1.6400000000000001,
|
329 |
+
"grad_norm": 3.906277656555176,
|
330 |
+
"learning_rate": 0.000118,
|
331 |
+
"loss": 0.3608,
|
332 |
+
"step": 410
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 1.6800000000000002,
|
336 |
+
"grad_norm": 2.591684103012085,
|
337 |
+
"learning_rate": 0.000116,
|
338 |
+
"loss": 0.4414,
|
339 |
+
"step": 420
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 1.72,
|
343 |
+
"grad_norm": 0.6823468804359436,
|
344 |
+
"learning_rate": 0.00011399999999999999,
|
345 |
+
"loss": 0.3937,
|
346 |
+
"step": 430
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 1.76,
|
350 |
+
"grad_norm": 2.4249002933502197,
|
351 |
+
"learning_rate": 0.00011200000000000001,
|
352 |
+
"loss": 0.2984,
|
353 |
+
"step": 440
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 1.8,
|
357 |
+
"grad_norm": 2.575287103652954,
|
358 |
+
"learning_rate": 0.00011000000000000002,
|
359 |
+
"loss": 0.4073,
|
360 |
+
"step": 450
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 1.8399999999999999,
|
364 |
+
"grad_norm": 0.8557507395744324,
|
365 |
+
"learning_rate": 0.00010800000000000001,
|
366 |
+
"loss": 0.3573,
|
367 |
+
"step": 460
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 1.88,
|
371 |
+
"grad_norm": 3.4324100017547607,
|
372 |
+
"learning_rate": 0.00010600000000000002,
|
373 |
+
"loss": 0.3758,
|
374 |
+
"step": 470
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 1.92,
|
378 |
+
"grad_norm": 2.3825552463531494,
|
379 |
+
"learning_rate": 0.00010400000000000001,
|
380 |
+
"loss": 0.2375,
|
381 |
+
"step": 480
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 1.96,
|
385 |
+
"grad_norm": 0.9951996207237244,
|
386 |
+
"learning_rate": 0.00010200000000000001,
|
387 |
+
"loss": 0.2496,
|
388 |
+
"step": 490
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 2.0,
|
392 |
+
"grad_norm": 2.2203187942504883,
|
393 |
+
"learning_rate": 0.0001,
|
394 |
+
"loss": 0.3279,
|
395 |
+
"step": 500
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 2.0,
|
399 |
+
"eval_accuracy": 0.876,
|
400 |
+
"eval_loss": 0.3772076666355133,
|
401 |
+
"eval_runtime": 14.2674,
|
402 |
+
"eval_samples_per_second": 70.09,
|
403 |
+
"eval_steps_per_second": 8.761,
|
404 |
+
"step": 500
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 2.04,
|
408 |
+
"grad_norm": 1.8857389688491821,
|
409 |
+
"learning_rate": 9.8e-05,
|
410 |
+
"loss": 0.1633,
|
411 |
+
"step": 510
|
412 |
+
},
|
413 |
+
{
|
414 |
+
"epoch": 2.08,
|
415 |
+
"grad_norm": 5.698770046234131,
|
416 |
+
"learning_rate": 9.6e-05,
|
417 |
+
"loss": 0.2812,
|
418 |
+
"step": 520
|
419 |
+
},
|
420 |
+
{
|
421 |
+
"epoch": 2.12,
|
422 |
+
"grad_norm": 1.7683120965957642,
|
423 |
+
"learning_rate": 9.4e-05,
|
424 |
+
"loss": 0.1895,
|
425 |
+
"step": 530
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 2.16,
|
429 |
+
"grad_norm": 0.6420239806175232,
|
430 |
+
"learning_rate": 9.200000000000001e-05,
|
431 |
+
"loss": 0.1732,
|
432 |
+
"step": 540
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 2.2,
|
436 |
+
"grad_norm": 0.8955737948417664,
|
437 |
+
"learning_rate": 9e-05,
|
438 |
+
"loss": 0.1557,
|
439 |
+
"step": 550
|
440 |
+
},
|
441 |
+
{
|
442 |
+
"epoch": 2.24,
|
443 |
+
"grad_norm": 2.202012300491333,
|
444 |
+
"learning_rate": 8.800000000000001e-05,
|
445 |
+
"loss": 0.2851,
|
446 |
+
"step": 560
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 2.2800000000000002,
|
450 |
+
"grad_norm": 3.6105308532714844,
|
451 |
+
"learning_rate": 8.6e-05,
|
452 |
+
"loss": 0.1645,
|
453 |
+
"step": 570
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 2.32,
|
457 |
+
"grad_norm": 3.514596462249756,
|
458 |
+
"learning_rate": 8.4e-05,
|
459 |
+
"loss": 0.1399,
|
460 |
+
"step": 580
|
461 |
+
},
|
462 |
+
{
|
463 |
+
"epoch": 2.36,
|
464 |
+
"grad_norm": 4.36515474319458,
|
465 |
+
"learning_rate": 8.2e-05,
|
466 |
+
"loss": 0.2495,
|
467 |
+
"step": 590
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 2.4,
|
471 |
+
"grad_norm": 0.10514427721500397,
|
472 |
+
"learning_rate": 8e-05,
|
473 |
+
"loss": 0.1337,
|
474 |
+
"step": 600
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 2.4,
|
478 |
+
"eval_accuracy": 0.869,
|
479 |
+
"eval_loss": 0.45184341073036194,
|
480 |
+
"eval_runtime": 14.683,
|
481 |
+
"eval_samples_per_second": 68.106,
|
482 |
+
"eval_steps_per_second": 8.513,
|
483 |
+
"step": 600
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 2.44,
|
487 |
+
"grad_norm": 1.140317440032959,
|
488 |
+
"learning_rate": 7.800000000000001e-05,
|
489 |
+
"loss": 0.1493,
|
490 |
+
"step": 610
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 2.48,
|
494 |
+
"grad_norm": 0.3709057569503784,
|
495 |
+
"learning_rate": 7.6e-05,
|
496 |
+
"loss": 0.164,
|
497 |
+
"step": 620
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 2.52,
|
501 |
+
"grad_norm": 3.097055196762085,
|
502 |
+
"learning_rate": 7.4e-05,
|
503 |
+
"loss": 0.208,
|
504 |
+
"step": 630
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 2.56,
|
508 |
+
"grad_norm": 3.960178852081299,
|
509 |
+
"learning_rate": 7.2e-05,
|
510 |
+
"loss": 0.2337,
|
511 |
+
"step": 640
|
512 |
+
},
|
513 |
+
{
|
514 |
+
"epoch": 2.6,
|
515 |
+
"grad_norm": 2.339881420135498,
|
516 |
+
"learning_rate": 7e-05,
|
517 |
+
"loss": 0.167,
|
518 |
+
"step": 650
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 2.64,
|
522 |
+
"grad_norm": 3.97763729095459,
|
523 |
+
"learning_rate": 6.800000000000001e-05,
|
524 |
+
"loss": 0.2026,
|
525 |
+
"step": 660
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 2.68,
|
529 |
+
"grad_norm": 0.5411188006401062,
|
530 |
+
"learning_rate": 6.6e-05,
|
531 |
+
"loss": 0.0895,
|
532 |
+
"step": 670
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 2.7199999999999998,
|
536 |
+
"grad_norm": 0.25824812054634094,
|
537 |
+
"learning_rate": 6.400000000000001e-05,
|
538 |
+
"loss": 0.0933,
|
539 |
+
"step": 680
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 2.76,
|
543 |
+
"grad_norm": 0.2557239830493927,
|
544 |
+
"learning_rate": 6.2e-05,
|
545 |
+
"loss": 0.1155,
|
546 |
+
"step": 690
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 2.8,
|
550 |
+
"grad_norm": 5.947152137756348,
|
551 |
+
"learning_rate": 6e-05,
|
552 |
+
"loss": 0.236,
|
553 |
+
"step": 700
|
554 |
+
},
|
555 |
+
{
|
556 |
+
"epoch": 2.8,
|
557 |
+
"eval_accuracy": 0.878,
|
558 |
+
"eval_loss": 0.37660717964172363,
|
559 |
+
"eval_runtime": 14.2296,
|
560 |
+
"eval_samples_per_second": 70.276,
|
561 |
+
"eval_steps_per_second": 8.785,
|
562 |
+
"step": 700
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 2.84,
|
566 |
+
"grad_norm": 2.6783535480499268,
|
567 |
+
"learning_rate": 5.8e-05,
|
568 |
+
"loss": 0.1437,
|
569 |
+
"step": 710
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 2.88,
|
573 |
+
"grad_norm": 1.7082568407058716,
|
574 |
+
"learning_rate": 5.6000000000000006e-05,
|
575 |
+
"loss": 0.1498,
|
576 |
+
"step": 720
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 2.92,
|
580 |
+
"grad_norm": 0.3654639720916748,
|
581 |
+
"learning_rate": 5.4000000000000005e-05,
|
582 |
+
"loss": 0.1544,
|
583 |
+
"step": 730
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 2.96,
|
587 |
+
"grad_norm": 2.7878735065460205,
|
588 |
+
"learning_rate": 5.2000000000000004e-05,
|
589 |
+
"loss": 0.2657,
|
590 |
+
"step": 740
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 3.0,
|
594 |
+
"grad_norm": 3.31339430809021,
|
595 |
+
"learning_rate": 5e-05,
|
596 |
+
"loss": 0.103,
|
597 |
+
"step": 750
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 3.04,
|
601 |
+
"grad_norm": 0.41359299421310425,
|
602 |
+
"learning_rate": 4.8e-05,
|
603 |
+
"loss": 0.0904,
|
604 |
+
"step": 760
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 3.08,
|
608 |
+
"grad_norm": 0.11081337183713913,
|
609 |
+
"learning_rate": 4.600000000000001e-05,
|
610 |
+
"loss": 0.0475,
|
611 |
+
"step": 770
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 3.12,
|
615 |
+
"grad_norm": 0.6292364001274109,
|
616 |
+
"learning_rate": 4.4000000000000006e-05,
|
617 |
+
"loss": 0.0613,
|
618 |
+
"step": 780
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 3.16,
|
622 |
+
"grad_norm": 0.06634623557329178,
|
623 |
+
"learning_rate": 4.2e-05,
|
624 |
+
"loss": 0.0419,
|
625 |
+
"step": 790
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 3.2,
|
629 |
+
"grad_norm": 3.720346212387085,
|
630 |
+
"learning_rate": 4e-05,
|
631 |
+
"loss": 0.0275,
|
632 |
+
"step": 800
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 3.2,
|
636 |
+
"eval_accuracy": 0.891,
|
637 |
+
"eval_loss": 0.3517528176307678,
|
638 |
+
"eval_runtime": 14.2729,
|
639 |
+
"eval_samples_per_second": 70.063,
|
640 |
+
"eval_steps_per_second": 8.758,
|
641 |
+
"step": 800
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 3.24,
|
645 |
+
"grad_norm": 0.15002816915512085,
|
646 |
+
"learning_rate": 3.8e-05,
|
647 |
+
"loss": 0.0425,
|
648 |
+
"step": 810
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 3.2800000000000002,
|
652 |
+
"grad_norm": 0.08299177885055542,
|
653 |
+
"learning_rate": 3.6e-05,
|
654 |
+
"loss": 0.0465,
|
655 |
+
"step": 820
|
656 |
+
},
|
657 |
+
{
|
658 |
+
"epoch": 3.32,
|
659 |
+
"grad_norm": 0.41334620118141174,
|
660 |
+
"learning_rate": 3.4000000000000007e-05,
|
661 |
+
"loss": 0.0434,
|
662 |
+
"step": 830
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 3.36,
|
666 |
+
"grad_norm": 0.5403936505317688,
|
667 |
+
"learning_rate": 3.2000000000000005e-05,
|
668 |
+
"loss": 0.0301,
|
669 |
+
"step": 840
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 3.4,
|
673 |
+
"grad_norm": 0.08261027932167053,
|
674 |
+
"learning_rate": 3e-05,
|
675 |
+
"loss": 0.072,
|
676 |
+
"step": 850
|
677 |
+
},
|
678 |
+
{
|
679 |
+
"epoch": 3.44,
|
680 |
+
"grad_norm": 1.0293442010879517,
|
681 |
+
"learning_rate": 2.8000000000000003e-05,
|
682 |
+
"loss": 0.082,
|
683 |
+
"step": 860
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 3.48,
|
687 |
+
"grad_norm": 1.7797234058380127,
|
688 |
+
"learning_rate": 2.6000000000000002e-05,
|
689 |
+
"loss": 0.0748,
|
690 |
+
"step": 870
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 3.52,
|
694 |
+
"grad_norm": 3.523738145828247,
|
695 |
+
"learning_rate": 2.4e-05,
|
696 |
+
"loss": 0.1751,
|
697 |
+
"step": 880
|
698 |
+
},
|
699 |
+
{
|
700 |
+
"epoch": 3.56,
|
701 |
+
"grad_norm": 0.06309465318918228,
|
702 |
+
"learning_rate": 2.2000000000000003e-05,
|
703 |
+
"loss": 0.0383,
|
704 |
+
"step": 890
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 3.6,
|
708 |
+
"grad_norm": 2.1426751613616943,
|
709 |
+
"learning_rate": 2e-05,
|
710 |
+
"loss": 0.0427,
|
711 |
+
"step": 900
|
712 |
+
},
|
713 |
+
{
|
714 |
+
"epoch": 3.6,
|
715 |
+
"eval_accuracy": 0.896,
|
716 |
+
"eval_loss": 0.3709311783313751,
|
717 |
+
"eval_runtime": 14.359,
|
718 |
+
"eval_samples_per_second": 69.643,
|
719 |
+
"eval_steps_per_second": 8.705,
|
720 |
+
"step": 900
|
721 |
+
},
|
722 |
+
{
|
723 |
+
"epoch": 3.64,
|
724 |
+
"grad_norm": 1.3229968547821045,
|
725 |
+
"learning_rate": 1.8e-05,
|
726 |
+
"loss": 0.0352,
|
727 |
+
"step": 910
|
728 |
+
},
|
729 |
+
{
|
730 |
+
"epoch": 3.68,
|
731 |
+
"grad_norm": 0.08263090997934341,
|
732 |
+
"learning_rate": 1.6000000000000003e-05,
|
733 |
+
"loss": 0.0192,
|
734 |
+
"step": 920
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 3.7199999999999998,
|
738 |
+
"grad_norm": 0.1414523720741272,
|
739 |
+
"learning_rate": 1.4000000000000001e-05,
|
740 |
+
"loss": 0.0724,
|
741 |
+
"step": 930
|
742 |
+
},
|
743 |
+
{
|
744 |
+
"epoch": 3.76,
|
745 |
+
"grad_norm": 0.05866268649697304,
|
746 |
+
"learning_rate": 1.2e-05,
|
747 |
+
"loss": 0.0289,
|
748 |
+
"step": 940
|
749 |
+
},
|
750 |
+
{
|
751 |
+
"epoch": 3.8,
|
752 |
+
"grad_norm": 0.08174656331539154,
|
753 |
+
"learning_rate": 1e-05,
|
754 |
+
"loss": 0.0264,
|
755 |
+
"step": 950
|
756 |
+
},
|
757 |
+
{
|
758 |
+
"epoch": 3.84,
|
759 |
+
"grad_norm": 0.07566811144351959,
|
760 |
+
"learning_rate": 8.000000000000001e-06,
|
761 |
+
"loss": 0.0225,
|
762 |
+
"step": 960
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 3.88,
|
766 |
+
"grad_norm": 0.06544584035873413,
|
767 |
+
"learning_rate": 6e-06,
|
768 |
+
"loss": 0.0488,
|
769 |
+
"step": 970
|
770 |
+
},
|
771 |
+
{
|
772 |
+
"epoch": 3.92,
|
773 |
+
"grad_norm": 0.2268047034740448,
|
774 |
+
"learning_rate": 4.000000000000001e-06,
|
775 |
+
"loss": 0.0423,
|
776 |
+
"step": 980
|
777 |
+
},
|
778 |
+
{
|
779 |
+
"epoch": 3.96,
|
780 |
+
"grad_norm": 0.05503053963184357,
|
781 |
+
"learning_rate": 2.0000000000000003e-06,
|
782 |
+
"loss": 0.0506,
|
783 |
+
"step": 990
|
784 |
+
},
|
785 |
+
{
|
786 |
+
"epoch": 4.0,
|
787 |
+
"grad_norm": 0.3238757252693176,
|
788 |
+
"learning_rate": 0.0,
|
789 |
+
"loss": 0.0363,
|
790 |
+
"step": 1000
|
791 |
+
},
|
792 |
+
{
|
793 |
+
"epoch": 4.0,
|
794 |
+
"eval_accuracy": 0.905,
|
795 |
+
"eval_loss": 0.34654033184051514,
|
796 |
+
"eval_runtime": 14.8487,
|
797 |
+
"eval_samples_per_second": 67.346,
|
798 |
+
"eval_steps_per_second": 8.418,
|
799 |
+
"step": 1000
|
800 |
+
},
|
801 |
+
{
|
802 |
+
"epoch": 4.0,
|
803 |
+
"step": 1000,
|
804 |
+
"total_flos": 1.239905171570688e+18,
|
805 |
+
"train_loss": 0.3179253642559052,
|
806 |
+
"train_runtime": 607.1163,
|
807 |
+
"train_samples_per_second": 26.354,
|
808 |
+
"train_steps_per_second": 1.647
|
809 |
+
}
|
810 |
+
],
|
811 |
+
"logging_steps": 10,
|
812 |
+
"max_steps": 1000,
|
813 |
+
"num_input_tokens_seen": 0,
|
814 |
+
"num_train_epochs": 4,
|
815 |
+
"save_steps": 100,
|
816 |
+
"stateful_callbacks": {
|
817 |
+
"TrainerControl": {
|
818 |
+
"args": {
|
819 |
+
"should_epoch_stop": false,
|
820 |
+
"should_evaluate": false,
|
821 |
+
"should_log": false,
|
822 |
+
"should_save": true,
|
823 |
+
"should_training_stop": true
|
824 |
+
},
|
825 |
+
"attributes": {}
|
826 |
+
}
|
827 |
+
},
|
828 |
+
"total_flos": 1.239905171570688e+18,
|
829 |
+
"train_batch_size": 16,
|
830 |
+
"trial_name": null,
|
831 |
+
"trial_params": null
|
832 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:50dacd679584c943310851e04614287d2d560396915ec76b9daca95419e88685
|
3 |
+
size 5112
|