Upload 8 files
Browse files- all_results.json +14 -0
- config.json +32 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- train_results.json +8 -0
- trainer_state.json +639 -0
- training_args.bin +3 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_f1": 0.963302752293578,
|
| 4 |
+
"eval_loss": 0.12418080866336823,
|
| 5 |
+
"eval_recall": 0.9156626506024096,
|
| 6 |
+
"eval_runtime": 10.2728,
|
| 7 |
+
"eval_samples_per_second": 41.177,
|
| 8 |
+
"eval_steps_per_second": 5.159,
|
| 9 |
+
"total_flos": 9.74539261393109e+17,
|
| 10 |
+
"train_loss": 0.13476083650806833,
|
| 11 |
+
"train_runtime": 481.1459,
|
| 12 |
+
"train_samples_per_second": 26.138,
|
| 13 |
+
"train_steps_per_second": 1.638
|
| 14 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "0",
|
| 13 |
+
"1": "1"
|
| 14 |
+
},
|
| 15 |
+
"image_size": 224,
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"0": "0",
|
| 20 |
+
"1": "1"
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-12,
|
| 23 |
+
"model_type": "vit",
|
| 24 |
+
"num_attention_heads": 12,
|
| 25 |
+
"num_channels": 3,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"patch_size": 16,
|
| 28 |
+
"problem_type": "single_label_classification",
|
| 29 |
+
"qkv_bias": true,
|
| 30 |
+
"torch_dtype": "float32",
|
| 31 |
+
"transformers_version": "4.38.2"
|
| 32 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_f1": 0.963302752293578,
|
| 4 |
+
"eval_loss": 0.12418080866336823,
|
| 5 |
+
"eval_runtime": 10.2728,
|
| 6 |
+
"eval_samples_per_second": 41.177,
|
| 7 |
+
"eval_steps_per_second": 5.159
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:126e9db7e36ade83203a205625d4994bca9edcec9df8367275922b824f950466
|
| 3 |
+
size 343223968
|
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 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 9.74539261393109e+17,
|
| 4 |
+
"train_loss": 0.13476083650806833,
|
| 5 |
+
"train_runtime": 481.1459,
|
| 6 |
+
"train_samples_per_second": 26.138,
|
| 7 |
+
"train_steps_per_second": 1.638
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,639 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.12418080866336823,
|
| 3 |
+
"best_model_checkpoint": "/content/drive/MyDrive/ml2/ML_project_2_course/weights1/checkpoint-700",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 788,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05,
|
| 13 |
+
"grad_norm": 1.3563565015792847,
|
| 14 |
+
"learning_rate": 0.00019746192893401016,
|
| 15 |
+
"loss": 0.6637,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1,
|
| 20 |
+
"grad_norm": 1.3717408180236816,
|
| 21 |
+
"learning_rate": 0.00019492385786802033,
|
| 22 |
+
"loss": 0.5792,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.15,
|
| 27 |
+
"grad_norm": 1.0099543333053589,
|
| 28 |
+
"learning_rate": 0.00019238578680203048,
|
| 29 |
+
"loss": 0.5253,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2,
|
| 34 |
+
"grad_norm": 5.920238494873047,
|
| 35 |
+
"learning_rate": 0.0001898477157360406,
|
| 36 |
+
"loss": 0.3762,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.25,
|
| 41 |
+
"grad_norm": 3.513320207595825,
|
| 42 |
+
"learning_rate": 0.00018730964467005077,
|
| 43 |
+
"loss": 0.3987,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.3,
|
| 48 |
+
"grad_norm": 1.2757742404937744,
|
| 49 |
+
"learning_rate": 0.00018477157360406092,
|
| 50 |
+
"loss": 0.3336,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.36,
|
| 55 |
+
"grad_norm": 1.0954958200454712,
|
| 56 |
+
"learning_rate": 0.00018223350253807107,
|
| 57 |
+
"loss": 0.4492,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.41,
|
| 62 |
+
"grad_norm": 1.1074856519699097,
|
| 63 |
+
"learning_rate": 0.00017969543147208124,
|
| 64 |
+
"loss": 0.3453,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.46,
|
| 69 |
+
"grad_norm": 2.97078800201416,
|
| 70 |
+
"learning_rate": 0.0001771573604060914,
|
| 71 |
+
"loss": 0.3103,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.51,
|
| 76 |
+
"grad_norm": 1.3375837802886963,
|
| 77 |
+
"learning_rate": 0.0001746192893401015,
|
| 78 |
+
"loss": 0.2034,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.51,
|
| 83 |
+
"eval_f1": 0.8307086614173228,
|
| 84 |
+
"eval_loss": 0.5720954537391663,
|
| 85 |
+
"eval_runtime": 5.4754,
|
| 86 |
+
"eval_samples_per_second": 77.255,
|
| 87 |
+
"eval_steps_per_second": 9.68,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.56,
|
| 92 |
+
"grad_norm": 1.1420457363128662,
|
| 93 |
+
"learning_rate": 0.00017208121827411168,
|
| 94 |
+
"loss": 0.3871,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.61,
|
| 99 |
+
"grad_norm": 1.5014958381652832,
|
| 100 |
+
"learning_rate": 0.00016954314720812183,
|
| 101 |
+
"loss": 0.242,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.66,
|
| 106 |
+
"grad_norm": 2.7499921321868896,
|
| 107 |
+
"learning_rate": 0.00016700507614213198,
|
| 108 |
+
"loss": 0.205,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.71,
|
| 113 |
+
"grad_norm": 1.3920105695724487,
|
| 114 |
+
"learning_rate": 0.00016446700507614215,
|
| 115 |
+
"loss": 0.2539,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.76,
|
| 120 |
+
"grad_norm": 2.2138984203338623,
|
| 121 |
+
"learning_rate": 0.0001619289340101523,
|
| 122 |
+
"loss": 0.3,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.81,
|
| 127 |
+
"grad_norm": 0.4511910080909729,
|
| 128 |
+
"learning_rate": 0.00015939086294416242,
|
| 129 |
+
"loss": 0.1906,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.86,
|
| 134 |
+
"grad_norm": 3.9145667552948,
|
| 135 |
+
"learning_rate": 0.0001568527918781726,
|
| 136 |
+
"loss": 0.2873,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.91,
|
| 141 |
+
"grad_norm": 7.277167797088623,
|
| 142 |
+
"learning_rate": 0.00015431472081218274,
|
| 143 |
+
"loss": 0.1775,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.96,
|
| 148 |
+
"grad_norm": 3.8661720752716064,
|
| 149 |
+
"learning_rate": 0.0001517766497461929,
|
| 150 |
+
"loss": 0.1682,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.02,
|
| 155 |
+
"grad_norm": 0.29035553336143494,
|
| 156 |
+
"learning_rate": 0.00014923857868020306,
|
| 157 |
+
"loss": 0.201,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.02,
|
| 162 |
+
"eval_f1": 0.9207048458149779,
|
| 163 |
+
"eval_loss": 0.2542865574359894,
|
| 164 |
+
"eval_runtime": 5.3252,
|
| 165 |
+
"eval_samples_per_second": 79.434,
|
| 166 |
+
"eval_steps_per_second": 9.953,
|
| 167 |
+
"step": 200
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 1.07,
|
| 171 |
+
"grad_norm": 7.454772472381592,
|
| 172 |
+
"learning_rate": 0.0001467005076142132,
|
| 173 |
+
"loss": 0.219,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.12,
|
| 178 |
+
"grad_norm": 2.4295766353607178,
|
| 179 |
+
"learning_rate": 0.00014416243654822336,
|
| 180 |
+
"loss": 0.1328,
|
| 181 |
+
"step": 220
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.17,
|
| 185 |
+
"grad_norm": 2.6288206577301025,
|
| 186 |
+
"learning_rate": 0.0001416243654822335,
|
| 187 |
+
"loss": 0.1695,
|
| 188 |
+
"step": 230
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.22,
|
| 192 |
+
"grad_norm": 1.8671033382415771,
|
| 193 |
+
"learning_rate": 0.00013908629441624365,
|
| 194 |
+
"loss": 0.1648,
|
| 195 |
+
"step": 240
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.27,
|
| 199 |
+
"grad_norm": 0.6234340667724609,
|
| 200 |
+
"learning_rate": 0.0001365482233502538,
|
| 201 |
+
"loss": 0.1318,
|
| 202 |
+
"step": 250
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 1.32,
|
| 206 |
+
"grad_norm": 1.7282599210739136,
|
| 207 |
+
"learning_rate": 0.00013401015228426397,
|
| 208 |
+
"loss": 0.1646,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.37,
|
| 213 |
+
"grad_norm": 8.562129020690918,
|
| 214 |
+
"learning_rate": 0.00013147208121827412,
|
| 215 |
+
"loss": 0.0765,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.42,
|
| 220 |
+
"grad_norm": 0.10765184462070465,
|
| 221 |
+
"learning_rate": 0.00012893401015228427,
|
| 222 |
+
"loss": 0.1424,
|
| 223 |
+
"step": 280
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 1.47,
|
| 227 |
+
"grad_norm": 1.785663366317749,
|
| 228 |
+
"learning_rate": 0.00012639593908629442,
|
| 229 |
+
"loss": 0.0838,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 1.52,
|
| 234 |
+
"grad_norm": 2.627047538757324,
|
| 235 |
+
"learning_rate": 0.00012385786802030456,
|
| 236 |
+
"loss": 0.2025,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.52,
|
| 241 |
+
"eval_f1": 0.9406392694063928,
|
| 242 |
+
"eval_loss": 0.21563898026943207,
|
| 243 |
+
"eval_runtime": 5.5211,
|
| 244 |
+
"eval_samples_per_second": 76.615,
|
| 245 |
+
"eval_steps_per_second": 9.599,
|
| 246 |
+
"step": 300
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.57,
|
| 250 |
+
"grad_norm": 0.36638501286506653,
|
| 251 |
+
"learning_rate": 0.00012131979695431472,
|
| 252 |
+
"loss": 0.1436,
|
| 253 |
+
"step": 310
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 1.62,
|
| 257 |
+
"grad_norm": 2.350069761276245,
|
| 258 |
+
"learning_rate": 0.00011878172588832489,
|
| 259 |
+
"loss": 0.0701,
|
| 260 |
+
"step": 320
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.68,
|
| 264 |
+
"grad_norm": 1.4287878274917603,
|
| 265 |
+
"learning_rate": 0.00011624365482233503,
|
| 266 |
+
"loss": 0.0843,
|
| 267 |
+
"step": 330
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.73,
|
| 271 |
+
"grad_norm": 0.32531386613845825,
|
| 272 |
+
"learning_rate": 0.0001137055837563452,
|
| 273 |
+
"loss": 0.1415,
|
| 274 |
+
"step": 340
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.78,
|
| 278 |
+
"grad_norm": 0.5714915990829468,
|
| 279 |
+
"learning_rate": 0.00011116751269035533,
|
| 280 |
+
"loss": 0.0751,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.83,
|
| 285 |
+
"grad_norm": 4.953071117401123,
|
| 286 |
+
"learning_rate": 0.00010862944162436547,
|
| 287 |
+
"loss": 0.126,
|
| 288 |
+
"step": 360
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.88,
|
| 292 |
+
"grad_norm": 4.382538318634033,
|
| 293 |
+
"learning_rate": 0.00010609137055837564,
|
| 294 |
+
"loss": 0.1279,
|
| 295 |
+
"step": 370
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.93,
|
| 299 |
+
"grad_norm": 2.4561574459075928,
|
| 300 |
+
"learning_rate": 0.0001035532994923858,
|
| 301 |
+
"loss": 0.0932,
|
| 302 |
+
"step": 380
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.98,
|
| 306 |
+
"grad_norm": 1.9325863122940063,
|
| 307 |
+
"learning_rate": 0.00010101522842639594,
|
| 308 |
+
"loss": 0.1361,
|
| 309 |
+
"step": 390
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 2.03,
|
| 313 |
+
"grad_norm": 0.1038457602262497,
|
| 314 |
+
"learning_rate": 9.847715736040609e-05,
|
| 315 |
+
"loss": 0.0564,
|
| 316 |
+
"step": 400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 2.03,
|
| 320 |
+
"eval_f1": 0.9452954048140045,
|
| 321 |
+
"eval_loss": 0.1903565376996994,
|
| 322 |
+
"eval_runtime": 6.5595,
|
| 323 |
+
"eval_samples_per_second": 64.486,
|
| 324 |
+
"eval_steps_per_second": 8.08,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 2.08,
|
| 329 |
+
"grad_norm": 0.12188208103179932,
|
| 330 |
+
"learning_rate": 9.593908629441625e-05,
|
| 331 |
+
"loss": 0.0947,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 2.13,
|
| 336 |
+
"grad_norm": 0.8643466234207153,
|
| 337 |
+
"learning_rate": 9.34010152284264e-05,
|
| 338 |
+
"loss": 0.058,
|
| 339 |
+
"step": 420
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.18,
|
| 343 |
+
"grad_norm": 0.06271246075630188,
|
| 344 |
+
"learning_rate": 9.086294416243655e-05,
|
| 345 |
+
"loss": 0.065,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 2.23,
|
| 350 |
+
"grad_norm": 0.593055784702301,
|
| 351 |
+
"learning_rate": 8.83248730964467e-05,
|
| 352 |
+
"loss": 0.0585,
|
| 353 |
+
"step": 440
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 2.28,
|
| 357 |
+
"grad_norm": 0.20611844956874847,
|
| 358 |
+
"learning_rate": 8.578680203045685e-05,
|
| 359 |
+
"loss": 0.0695,
|
| 360 |
+
"step": 450
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 2.34,
|
| 364 |
+
"grad_norm": 0.09795264154672623,
|
| 365 |
+
"learning_rate": 8.324873096446701e-05,
|
| 366 |
+
"loss": 0.0351,
|
| 367 |
+
"step": 460
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 2.39,
|
| 371 |
+
"grad_norm": 0.04693546146154404,
|
| 372 |
+
"learning_rate": 8.071065989847716e-05,
|
| 373 |
+
"loss": 0.0367,
|
| 374 |
+
"step": 470
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.44,
|
| 378 |
+
"grad_norm": 0.09200535714626312,
|
| 379 |
+
"learning_rate": 7.817258883248731e-05,
|
| 380 |
+
"loss": 0.0266,
|
| 381 |
+
"step": 480
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 2.49,
|
| 385 |
+
"grad_norm": 0.134303480386734,
|
| 386 |
+
"learning_rate": 7.563451776649747e-05,
|
| 387 |
+
"loss": 0.0306,
|
| 388 |
+
"step": 490
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 2.54,
|
| 392 |
+
"grad_norm": 0.041602715849876404,
|
| 393 |
+
"learning_rate": 7.309644670050762e-05,
|
| 394 |
+
"loss": 0.0262,
|
| 395 |
+
"step": 500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 2.54,
|
| 399 |
+
"eval_f1": 0.9495412844036697,
|
| 400 |
+
"eval_loss": 0.21375472843647003,
|
| 401 |
+
"eval_runtime": 5.9305,
|
| 402 |
+
"eval_samples_per_second": 71.326,
|
| 403 |
+
"eval_steps_per_second": 8.937,
|
| 404 |
+
"step": 500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 2.59,
|
| 408 |
+
"grad_norm": 5.353679656982422,
|
| 409 |
+
"learning_rate": 7.055837563451776e-05,
|
| 410 |
+
"loss": 0.0615,
|
| 411 |
+
"step": 510
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 2.64,
|
| 415 |
+
"grad_norm": 7.85841703414917,
|
| 416 |
+
"learning_rate": 6.802030456852793e-05,
|
| 417 |
+
"loss": 0.0927,
|
| 418 |
+
"step": 520
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 2.69,
|
| 422 |
+
"grad_norm": 0.5664938688278198,
|
| 423 |
+
"learning_rate": 6.548223350253807e-05,
|
| 424 |
+
"loss": 0.1466,
|
| 425 |
+
"step": 530
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 2.74,
|
| 429 |
+
"grad_norm": 0.04644864797592163,
|
| 430 |
+
"learning_rate": 6.294416243654822e-05,
|
| 431 |
+
"loss": 0.0148,
|
| 432 |
+
"step": 540
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 2.79,
|
| 436 |
+
"grad_norm": 0.043112464249134064,
|
| 437 |
+
"learning_rate": 6.040609137055838e-05,
|
| 438 |
+
"loss": 0.041,
|
| 439 |
+
"step": 550
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 2.84,
|
| 443 |
+
"grad_norm": 0.9218811392784119,
|
| 444 |
+
"learning_rate": 5.786802030456853e-05,
|
| 445 |
+
"loss": 0.0998,
|
| 446 |
+
"step": 560
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 2.89,
|
| 450 |
+
"grad_norm": 0.09559272974729538,
|
| 451 |
+
"learning_rate": 5.532994923857868e-05,
|
| 452 |
+
"loss": 0.0447,
|
| 453 |
+
"step": 570
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 2.94,
|
| 457 |
+
"grad_norm": 0.04861769080162048,
|
| 458 |
+
"learning_rate": 5.2791878172588836e-05,
|
| 459 |
+
"loss": 0.0296,
|
| 460 |
+
"step": 580
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 2.99,
|
| 464 |
+
"grad_norm": 0.13610342144966125,
|
| 465 |
+
"learning_rate": 5.0253807106598984e-05,
|
| 466 |
+
"loss": 0.037,
|
| 467 |
+
"step": 590
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 3.05,
|
| 471 |
+
"grad_norm": 0.048563599586486816,
|
| 472 |
+
"learning_rate": 4.771573604060914e-05,
|
| 473 |
+
"loss": 0.0222,
|
| 474 |
+
"step": 600
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 3.05,
|
| 478 |
+
"eval_f1": 0.9508196721311475,
|
| 479 |
+
"eval_loss": 0.15546846389770508,
|
| 480 |
+
"eval_runtime": 5.4832,
|
| 481 |
+
"eval_samples_per_second": 77.145,
|
| 482 |
+
"eval_steps_per_second": 9.666,
|
| 483 |
+
"step": 600
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 3.1,
|
| 487 |
+
"grad_norm": 0.05200362578034401,
|
| 488 |
+
"learning_rate": 4.517766497461929e-05,
|
| 489 |
+
"loss": 0.0118,
|
| 490 |
+
"step": 610
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 3.15,
|
| 494 |
+
"grad_norm": 1.2061017751693726,
|
| 495 |
+
"learning_rate": 4.2639593908629446e-05,
|
| 496 |
+
"loss": 0.0469,
|
| 497 |
+
"step": 620
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 3.2,
|
| 501 |
+
"grad_norm": 0.05518200248479843,
|
| 502 |
+
"learning_rate": 4.010152284263959e-05,
|
| 503 |
+
"loss": 0.0228,
|
| 504 |
+
"step": 630
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 3.25,
|
| 508 |
+
"grad_norm": 0.19791868329048157,
|
| 509 |
+
"learning_rate": 3.756345177664975e-05,
|
| 510 |
+
"loss": 0.0184,
|
| 511 |
+
"step": 640
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 3.3,
|
| 515 |
+
"grad_norm": 0.10160644352436066,
|
| 516 |
+
"learning_rate": 3.50253807106599e-05,
|
| 517 |
+
"loss": 0.0267,
|
| 518 |
+
"step": 650
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 3.35,
|
| 522 |
+
"grad_norm": 0.04751751944422722,
|
| 523 |
+
"learning_rate": 3.248730964467005e-05,
|
| 524 |
+
"loss": 0.0096,
|
| 525 |
+
"step": 660
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 3.4,
|
| 529 |
+
"grad_norm": 0.04644055664539337,
|
| 530 |
+
"learning_rate": 2.9949238578680206e-05,
|
| 531 |
+
"loss": 0.0367,
|
| 532 |
+
"step": 670
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 3.45,
|
| 536 |
+
"grad_norm": 0.060281168669462204,
|
| 537 |
+
"learning_rate": 2.7411167512690357e-05,
|
| 538 |
+
"loss": 0.0068,
|
| 539 |
+
"step": 680
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 3.5,
|
| 543 |
+
"grad_norm": 0.034235160797834396,
|
| 544 |
+
"learning_rate": 2.4873096446700507e-05,
|
| 545 |
+
"loss": 0.0398,
|
| 546 |
+
"step": 690
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 3.55,
|
| 550 |
+
"grad_norm": 0.3630073368549347,
|
| 551 |
+
"learning_rate": 2.233502538071066e-05,
|
| 552 |
+
"loss": 0.0069,
|
| 553 |
+
"step": 700
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 3.55,
|
| 557 |
+
"eval_f1": 0.963302752293578,
|
| 558 |
+
"eval_loss": 0.12418080866336823,
|
| 559 |
+
"eval_runtime": 6.5914,
|
| 560 |
+
"eval_samples_per_second": 64.175,
|
| 561 |
+
"eval_steps_per_second": 8.041,
|
| 562 |
+
"step": 700
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 3.6,
|
| 566 |
+
"grad_norm": 0.040880296379327774,
|
| 567 |
+
"learning_rate": 1.9796954314720815e-05,
|
| 568 |
+
"loss": 0.0045,
|
| 569 |
+
"step": 710
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 3.65,
|
| 573 |
+
"grad_norm": 0.5524014830589294,
|
| 574 |
+
"learning_rate": 1.7258883248730966e-05,
|
| 575 |
+
"loss": 0.04,
|
| 576 |
+
"step": 720
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 3.71,
|
| 580 |
+
"grad_norm": 0.03459872677922249,
|
| 581 |
+
"learning_rate": 1.4720812182741117e-05,
|
| 582 |
+
"loss": 0.0131,
|
| 583 |
+
"step": 730
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 3.76,
|
| 587 |
+
"grad_norm": 2.4353725910186768,
|
| 588 |
+
"learning_rate": 1.218274111675127e-05,
|
| 589 |
+
"loss": 0.0114,
|
| 590 |
+
"step": 740
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 3.81,
|
| 594 |
+
"grad_norm": 0.049515530467033386,
|
| 595 |
+
"learning_rate": 9.644670050761421e-06,
|
| 596 |
+
"loss": 0.0163,
|
| 597 |
+
"step": 750
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 3.86,
|
| 601 |
+
"grad_norm": 0.033101875334978104,
|
| 602 |
+
"learning_rate": 7.106598984771575e-06,
|
| 603 |
+
"loss": 0.0108,
|
| 604 |
+
"step": 760
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 3.91,
|
| 608 |
+
"grad_norm": 0.08629802614450455,
|
| 609 |
+
"learning_rate": 4.568527918781726e-06,
|
| 610 |
+
"loss": 0.0435,
|
| 611 |
+
"step": 770
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 3.96,
|
| 615 |
+
"grad_norm": 0.09244952350854874,
|
| 616 |
+
"learning_rate": 2.030456852791878e-06,
|
| 617 |
+
"loss": 0.0195,
|
| 618 |
+
"step": 780
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 4.0,
|
| 622 |
+
"step": 788,
|
| 623 |
+
"total_flos": 9.74539261393109e+17,
|
| 624 |
+
"train_loss": 0.13476083650806833,
|
| 625 |
+
"train_runtime": 481.1459,
|
| 626 |
+
"train_samples_per_second": 26.138,
|
| 627 |
+
"train_steps_per_second": 1.638
|
| 628 |
+
}
|
| 629 |
+
],
|
| 630 |
+
"logging_steps": 10,
|
| 631 |
+
"max_steps": 788,
|
| 632 |
+
"num_input_tokens_seen": 0,
|
| 633 |
+
"num_train_epochs": 4,
|
| 634 |
+
"save_steps": 100,
|
| 635 |
+
"total_flos": 9.74539261393109e+17,
|
| 636 |
+
"train_batch_size": 16,
|
| 637 |
+
"trial_name": null,
|
| 638 |
+
"trial_params": null
|
| 639 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1305a0c51f06f8299442f21d8dd69f9718af8c622266c7e7ebfc4cce313dea04
|
| 3 |
+
size 4984
|