Upload 8 files
Browse files- all_results.json +13 -0
- config.json +58 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +36 -0
- train_results.json +8 -0
- trainer_state.json +1594 -0
- training_args.bin +3 -0
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 96.20253164556962,
|
| 3 |
+
"eval_accuracy": 0.8746031746031746,
|
| 4 |
+
"eval_loss": 0.5267017483711243,
|
| 5 |
+
"eval_runtime": 10.7929,
|
| 6 |
+
"eval_samples_per_second": 116.743,
|
| 7 |
+
"eval_steps_per_second": 0.927,
|
| 8 |
+
"total_flos": 7.515490775048022e+19,
|
| 9 |
+
"train_loss": 0.33647052476280614,
|
| 10 |
+
"train_runtime": 20573.1873,
|
| 11 |
+
"train_samples_per_second": 48.996,
|
| 12 |
+
"train_steps_per_second": 0.092
|
| 13 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch16-224",
|
| 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": "sitting",
|
| 13 |
+
"1": "using_laptop",
|
| 14 |
+
"2": "hugging",
|
| 15 |
+
"3": "sleeping",
|
| 16 |
+
"4": "drinking",
|
| 17 |
+
"5": "clapping",
|
| 18 |
+
"6": "dancing",
|
| 19 |
+
"7": "cycling",
|
| 20 |
+
"8": "calling",
|
| 21 |
+
"9": "laughing",
|
| 22 |
+
"10": "eating",
|
| 23 |
+
"11": "fighting",
|
| 24 |
+
"12": "listening_to_music",
|
| 25 |
+
"13": "running",
|
| 26 |
+
"14": "texting"
|
| 27 |
+
},
|
| 28 |
+
"image_size": 224,
|
| 29 |
+
"initializer_range": 0.02,
|
| 30 |
+
"intermediate_size": 3072,
|
| 31 |
+
"label2id": {
|
| 32 |
+
"calling": 8,
|
| 33 |
+
"clapping": 5,
|
| 34 |
+
"cycling": 7,
|
| 35 |
+
"dancing": 6,
|
| 36 |
+
"drinking": 4,
|
| 37 |
+
"eating": 10,
|
| 38 |
+
"fighting": 11,
|
| 39 |
+
"hugging": 2,
|
| 40 |
+
"laughing": 9,
|
| 41 |
+
"listening_to_music": 12,
|
| 42 |
+
"running": 13,
|
| 43 |
+
"sitting": 0,
|
| 44 |
+
"sleeping": 3,
|
| 45 |
+
"texting": 14,
|
| 46 |
+
"using_laptop": 1
|
| 47 |
+
},
|
| 48 |
+
"layer_norm_eps": 1e-12,
|
| 49 |
+
"model_type": "vit",
|
| 50 |
+
"num_attention_heads": 12,
|
| 51 |
+
"num_channels": 3,
|
| 52 |
+
"num_hidden_layers": 12,
|
| 53 |
+
"patch_size": 16,
|
| 54 |
+
"problem_type": "single_label_classification",
|
| 55 |
+
"qkv_bias": true,
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.41.2"
|
| 58 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 96.20253164556962,
|
| 3 |
+
"eval_accuracy": 0.8746031746031746,
|
| 4 |
+
"eval_loss": 0.5267017483711243,
|
| 5 |
+
"eval_runtime": 10.7929,
|
| 6 |
+
"eval_samples_per_second": 116.743,
|
| 7 |
+
"eval_steps_per_second": 0.927
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff64402a55469a315d1e1c5a2136d2f36fa2972e9e72b454371736d9368e64da
|
| 3 |
+
size 343263964
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"images",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_rescale",
|
| 8 |
+
"rescale_factor",
|
| 9 |
+
"do_normalize",
|
| 10 |
+
"image_mean",
|
| 11 |
+
"image_std",
|
| 12 |
+
"return_tensors",
|
| 13 |
+
"data_format",
|
| 14 |
+
"input_data_format"
|
| 15 |
+
],
|
| 16 |
+
"do_normalize": true,
|
| 17 |
+
"do_rescale": true,
|
| 18 |
+
"do_resize": true,
|
| 19 |
+
"image_mean": [
|
| 20 |
+
0.5,
|
| 21 |
+
0.5,
|
| 22 |
+
0.5
|
| 23 |
+
],
|
| 24 |
+
"image_processor_type": "ViTImageProcessor",
|
| 25 |
+
"image_std": [
|
| 26 |
+
0.5,
|
| 27 |
+
0.5,
|
| 28 |
+
0.5
|
| 29 |
+
],
|
| 30 |
+
"resample": 2,
|
| 31 |
+
"rescale_factor": 0.00392156862745098,
|
| 32 |
+
"size": {
|
| 33 |
+
"height": 224,
|
| 34 |
+
"width": 224
|
| 35 |
+
}
|
| 36 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 96.20253164556962,
|
| 3 |
+
"total_flos": 7.515490775048022e+19,
|
| 4 |
+
"train_loss": 0.33647052476280614,
|
| 5 |
+
"train_runtime": 20573.1873,
|
| 6 |
+
"train_samples_per_second": 48.996,
|
| 7 |
+
"train_steps_per_second": 0.092
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1594 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8746031746031746,
|
| 3 |
+
"best_model_checkpoint": "CP2_HAR_vit-base-patch16-224/checkpoint-1382",
|
| 4 |
+
"epoch": 96.20253164556962,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1900,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.9620253164556962,
|
| 13 |
+
"grad_norm": 2.8217339515686035,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 2.8429,
|
| 16 |
+
"step": 19
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.9620253164556962,
|
| 20 |
+
"eval_accuracy": 0.16904761904761906,
|
| 21 |
+
"eval_loss": 2.6474363803863525,
|
| 22 |
+
"eval_runtime": 10.7675,
|
| 23 |
+
"eval_samples_per_second": 117.018,
|
| 24 |
+
"eval_steps_per_second": 0.929,
|
| 25 |
+
"step": 19
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.9746835443037973,
|
| 29 |
+
"grad_norm": 2.534130334854126,
|
| 30 |
+
"learning_rate": 1.0263157894736843e-05,
|
| 31 |
+
"loss": 2.5152,
|
| 32 |
+
"step": 39
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.9746835443037973,
|
| 36 |
+
"eval_accuracy": 0.31587301587301586,
|
| 37 |
+
"eval_loss": 2.342618227005005,
|
| 38 |
+
"eval_runtime": 10.7708,
|
| 39 |
+
"eval_samples_per_second": 116.983,
|
| 40 |
+
"eval_steps_per_second": 0.928,
|
| 41 |
+
"step": 39
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 2.9873417721518987,
|
| 45 |
+
"grad_norm": 2.3139591217041016,
|
| 46 |
+
"learning_rate": 1.5526315789473686e-05,
|
| 47 |
+
"loss": 2.1548,
|
| 48 |
+
"step": 59
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 2.9873417721518987,
|
| 52 |
+
"eval_accuracy": 0.5626984126984127,
|
| 53 |
+
"eval_loss": 1.8545457124710083,
|
| 54 |
+
"eval_runtime": 10.8474,
|
| 55 |
+
"eval_samples_per_second": 116.157,
|
| 56 |
+
"eval_steps_per_second": 0.922,
|
| 57 |
+
"step": 59
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.0,
|
| 61 |
+
"grad_norm": 1.9620369672775269,
|
| 62 |
+
"learning_rate": 2.078947368421053e-05,
|
| 63 |
+
"loss": 1.6569,
|
| 64 |
+
"step": 79
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 4.0,
|
| 68 |
+
"eval_accuracy": 0.7261904761904762,
|
| 69 |
+
"eval_loss": 1.2644377946853638,
|
| 70 |
+
"eval_runtime": 10.7977,
|
| 71 |
+
"eval_samples_per_second": 116.692,
|
| 72 |
+
"eval_steps_per_second": 0.926,
|
| 73 |
+
"step": 79
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 4.962025316455696,
|
| 77 |
+
"grad_norm": 1.4392253160476685,
|
| 78 |
+
"learning_rate": 2.578947368421053e-05,
|
| 79 |
+
"loss": 1.2393,
|
| 80 |
+
"step": 98
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 4.962025316455696,
|
| 84 |
+
"eval_accuracy": 0.7714285714285715,
|
| 85 |
+
"eval_loss": 0.8716472387313843,
|
| 86 |
+
"eval_runtime": 10.7696,
|
| 87 |
+
"eval_samples_per_second": 116.996,
|
| 88 |
+
"eval_steps_per_second": 0.929,
|
| 89 |
+
"step": 98
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 5.974683544303797,
|
| 93 |
+
"grad_norm": 1.3771088123321533,
|
| 94 |
+
"learning_rate": 3.105263157894737e-05,
|
| 95 |
+
"loss": 0.8982,
|
| 96 |
+
"step": 118
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 5.974683544303797,
|
| 100 |
+
"eval_accuracy": 0.8150793650793651,
|
| 101 |
+
"eval_loss": 0.6652108430862427,
|
| 102 |
+
"eval_runtime": 10.7233,
|
| 103 |
+
"eval_samples_per_second": 117.501,
|
| 104 |
+
"eval_steps_per_second": 0.933,
|
| 105 |
+
"step": 118
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 6.987341772151899,
|
| 109 |
+
"grad_norm": 1.6404207944869995,
|
| 110 |
+
"learning_rate": 3.6315789473684214e-05,
|
| 111 |
+
"loss": 0.7694,
|
| 112 |
+
"step": 138
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 6.987341772151899,
|
| 116 |
+
"eval_accuracy": 0.830952380952381,
|
| 117 |
+
"eval_loss": 0.596939206123352,
|
| 118 |
+
"eval_runtime": 10.7687,
|
| 119 |
+
"eval_samples_per_second": 117.006,
|
| 120 |
+
"eval_steps_per_second": 0.929,
|
| 121 |
+
"step": 138
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 8.0,
|
| 125 |
+
"grad_norm": 1.4968894720077515,
|
| 126 |
+
"learning_rate": 4.157894736842106e-05,
|
| 127 |
+
"loss": 0.6819,
|
| 128 |
+
"step": 158
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.0,
|
| 132 |
+
"eval_accuracy": 0.8396825396825397,
|
| 133 |
+
"eval_loss": 0.5484516620635986,
|
| 134 |
+
"eval_runtime": 10.8135,
|
| 135 |
+
"eval_samples_per_second": 116.521,
|
| 136 |
+
"eval_steps_per_second": 0.925,
|
| 137 |
+
"step": 158
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 8.962025316455696,
|
| 141 |
+
"grad_norm": 1.414362907409668,
|
| 142 |
+
"learning_rate": 4.657894736842106e-05,
|
| 143 |
+
"loss": 0.6628,
|
| 144 |
+
"step": 177
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 8.962025316455696,
|
| 148 |
+
"eval_accuracy": 0.8476190476190476,
|
| 149 |
+
"eval_loss": 0.5053515434265137,
|
| 150 |
+
"eval_runtime": 10.7521,
|
| 151 |
+
"eval_samples_per_second": 117.186,
|
| 152 |
+
"eval_steps_per_second": 0.93,
|
| 153 |
+
"step": 177
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 9.974683544303797,
|
| 157 |
+
"grad_norm": 1.5052249431610107,
|
| 158 |
+
"learning_rate": 4.97953216374269e-05,
|
| 159 |
+
"loss": 0.5759,
|
| 160 |
+
"step": 197
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 9.974683544303797,
|
| 164 |
+
"eval_accuracy": 0.8476190476190476,
|
| 165 |
+
"eval_loss": 0.5064724683761597,
|
| 166 |
+
"eval_runtime": 10.8275,
|
| 167 |
+
"eval_samples_per_second": 116.37,
|
| 168 |
+
"eval_steps_per_second": 0.924,
|
| 169 |
+
"step": 197
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 10.987341772151899,
|
| 173 |
+
"grad_norm": 1.4250831604003906,
|
| 174 |
+
"learning_rate": 4.921052631578947e-05,
|
| 175 |
+
"loss": 0.5385,
|
| 176 |
+
"step": 217
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 10.987341772151899,
|
| 180 |
+
"eval_accuracy": 0.8420634920634921,
|
| 181 |
+
"eval_loss": 0.4821438789367676,
|
| 182 |
+
"eval_runtime": 10.8099,
|
| 183 |
+
"eval_samples_per_second": 116.56,
|
| 184 |
+
"eval_steps_per_second": 0.925,
|
| 185 |
+
"step": 217
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 12.0,
|
| 189 |
+
"grad_norm": 1.3822650909423828,
|
| 190 |
+
"learning_rate": 4.8625730994152046e-05,
|
| 191 |
+
"loss": 0.5022,
|
| 192 |
+
"step": 237
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 12.0,
|
| 196 |
+
"eval_accuracy": 0.8507936507936508,
|
| 197 |
+
"eval_loss": 0.47235107421875,
|
| 198 |
+
"eval_runtime": 10.7944,
|
| 199 |
+
"eval_samples_per_second": 116.727,
|
| 200 |
+
"eval_steps_per_second": 0.926,
|
| 201 |
+
"step": 237
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 12.962025316455696,
|
| 205 |
+
"grad_norm": 1.543864369392395,
|
| 206 |
+
"learning_rate": 4.807017543859649e-05,
|
| 207 |
+
"loss": 0.4841,
|
| 208 |
+
"step": 256
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 12.962025316455696,
|
| 212 |
+
"eval_accuracy": 0.8587301587301587,
|
| 213 |
+
"eval_loss": 0.48088887333869934,
|
| 214 |
+
"eval_runtime": 10.7741,
|
| 215 |
+
"eval_samples_per_second": 116.947,
|
| 216 |
+
"eval_steps_per_second": 0.928,
|
| 217 |
+
"step": 256
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 13.974683544303797,
|
| 221 |
+
"grad_norm": 1.4722260236740112,
|
| 222 |
+
"learning_rate": 4.7485380116959065e-05,
|
| 223 |
+
"loss": 0.4543,
|
| 224 |
+
"step": 276
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 13.974683544303797,
|
| 228 |
+
"eval_accuracy": 0.861904761904762,
|
| 229 |
+
"eval_loss": 0.4476229250431061,
|
| 230 |
+
"eval_runtime": 10.73,
|
| 231 |
+
"eval_samples_per_second": 117.428,
|
| 232 |
+
"eval_steps_per_second": 0.932,
|
| 233 |
+
"step": 276
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 14.987341772151899,
|
| 237 |
+
"grad_norm": 1.5065507888793945,
|
| 238 |
+
"learning_rate": 4.690058479532164e-05,
|
| 239 |
+
"loss": 0.4356,
|
| 240 |
+
"step": 296
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 14.987341772151899,
|
| 244 |
+
"eval_accuracy": 0.8579365079365079,
|
| 245 |
+
"eval_loss": 0.47357481718063354,
|
| 246 |
+
"eval_runtime": 10.7482,
|
| 247 |
+
"eval_samples_per_second": 117.229,
|
| 248 |
+
"eval_steps_per_second": 0.93,
|
| 249 |
+
"step": 296
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 16.0,
|
| 253 |
+
"grad_norm": 1.6331088542938232,
|
| 254 |
+
"learning_rate": 4.6315789473684214e-05,
|
| 255 |
+
"loss": 0.4021,
|
| 256 |
+
"step": 316
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 16.0,
|
| 260 |
+
"eval_accuracy": 0.8587301587301587,
|
| 261 |
+
"eval_loss": 0.46398431062698364,
|
| 262 |
+
"eval_runtime": 10.7745,
|
| 263 |
+
"eval_samples_per_second": 116.943,
|
| 264 |
+
"eval_steps_per_second": 0.928,
|
| 265 |
+
"step": 316
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 16.962025316455698,
|
| 269 |
+
"grad_norm": 1.7429494857788086,
|
| 270 |
+
"learning_rate": 4.576023391812866e-05,
|
| 271 |
+
"loss": 0.4073,
|
| 272 |
+
"step": 335
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 16.962025316455698,
|
| 276 |
+
"eval_accuracy": 0.8579365079365079,
|
| 277 |
+
"eval_loss": 0.4629625976085663,
|
| 278 |
+
"eval_runtime": 10.7423,
|
| 279 |
+
"eval_samples_per_second": 117.293,
|
| 280 |
+
"eval_steps_per_second": 0.931,
|
| 281 |
+
"step": 335
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 17.974683544303797,
|
| 285 |
+
"grad_norm": 1.3264607191085815,
|
| 286 |
+
"learning_rate": 4.517543859649123e-05,
|
| 287 |
+
"loss": 0.3782,
|
| 288 |
+
"step": 355
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 17.974683544303797,
|
| 292 |
+
"eval_accuracy": 0.8595238095238096,
|
| 293 |
+
"eval_loss": 0.4655977785587311,
|
| 294 |
+
"eval_runtime": 10.8299,
|
| 295 |
+
"eval_samples_per_second": 116.344,
|
| 296 |
+
"eval_steps_per_second": 0.923,
|
| 297 |
+
"step": 355
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 18.9873417721519,
|
| 301 |
+
"grad_norm": 1.481920599937439,
|
| 302 |
+
"learning_rate": 4.4590643274853806e-05,
|
| 303 |
+
"loss": 0.3617,
|
| 304 |
+
"step": 375
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 18.9873417721519,
|
| 308 |
+
"eval_accuracy": 0.861904761904762,
|
| 309 |
+
"eval_loss": 0.4484713077545166,
|
| 310 |
+
"eval_runtime": 10.749,
|
| 311 |
+
"eval_samples_per_second": 117.221,
|
| 312 |
+
"eval_steps_per_second": 0.93,
|
| 313 |
+
"step": 375
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 20.0,
|
| 317 |
+
"grad_norm": 1.3743647336959839,
|
| 318 |
+
"learning_rate": 4.400584795321638e-05,
|
| 319 |
+
"loss": 0.3448,
|
| 320 |
+
"step": 395
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 20.0,
|
| 324 |
+
"eval_accuracy": 0.861904761904762,
|
| 325 |
+
"eval_loss": 0.4735279381275177,
|
| 326 |
+
"eval_runtime": 10.7186,
|
| 327 |
+
"eval_samples_per_second": 117.553,
|
| 328 |
+
"eval_steps_per_second": 0.933,
|
| 329 |
+
"step": 395
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 20.962025316455698,
|
| 333 |
+
"grad_norm": 1.3664813041687012,
|
| 334 |
+
"learning_rate": 4.345029239766082e-05,
|
| 335 |
+
"loss": 0.3549,
|
| 336 |
+
"step": 414
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 20.962025316455698,
|
| 340 |
+
"eval_accuracy": 0.8571428571428571,
|
| 341 |
+
"eval_loss": 0.4780659079551697,
|
| 342 |
+
"eval_runtime": 10.7555,
|
| 343 |
+
"eval_samples_per_second": 117.149,
|
| 344 |
+
"eval_steps_per_second": 0.93,
|
| 345 |
+
"step": 414
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 21.974683544303797,
|
| 349 |
+
"grad_norm": 1.4020764827728271,
|
| 350 |
+
"learning_rate": 4.286549707602339e-05,
|
| 351 |
+
"loss": 0.3195,
|
| 352 |
+
"step": 434
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 21.974683544303797,
|
| 356 |
+
"eval_accuracy": 0.861904761904762,
|
| 357 |
+
"eval_loss": 0.4818320572376251,
|
| 358 |
+
"eval_runtime": 10.6974,
|
| 359 |
+
"eval_samples_per_second": 117.785,
|
| 360 |
+
"eval_steps_per_second": 0.935,
|
| 361 |
+
"step": 434
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 22.9873417721519,
|
| 365 |
+
"grad_norm": 1.2878130674362183,
|
| 366 |
+
"learning_rate": 4.228070175438597e-05,
|
| 367 |
+
"loss": 0.3219,
|
| 368 |
+
"step": 454
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 22.9873417721519,
|
| 372 |
+
"eval_accuracy": 0.8650793650793651,
|
| 373 |
+
"eval_loss": 0.47401970624923706,
|
| 374 |
+
"eval_runtime": 10.7479,
|
| 375 |
+
"eval_samples_per_second": 117.232,
|
| 376 |
+
"eval_steps_per_second": 0.93,
|
| 377 |
+
"step": 454
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 24.0,
|
| 381 |
+
"grad_norm": 1.5816872119903564,
|
| 382 |
+
"learning_rate": 4.169590643274854e-05,
|
| 383 |
+
"loss": 0.2966,
|
| 384 |
+
"step": 474
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 24.0,
|
| 388 |
+
"eval_accuracy": 0.8642857142857143,
|
| 389 |
+
"eval_loss": 0.4857538044452667,
|
| 390 |
+
"eval_runtime": 10.884,
|
| 391 |
+
"eval_samples_per_second": 115.766,
|
| 392 |
+
"eval_steps_per_second": 0.919,
|
| 393 |
+
"step": 474
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 24.962025316455698,
|
| 397 |
+
"grad_norm": 1.4161866903305054,
|
| 398 |
+
"learning_rate": 4.1140350877192985e-05,
|
| 399 |
+
"loss": 0.322,
|
| 400 |
+
"step": 493
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 24.962025316455698,
|
| 404 |
+
"eval_accuracy": 0.8579365079365079,
|
| 405 |
+
"eval_loss": 0.4993390738964081,
|
| 406 |
+
"eval_runtime": 10.7563,
|
| 407 |
+
"eval_samples_per_second": 117.141,
|
| 408 |
+
"eval_steps_per_second": 0.93,
|
| 409 |
+
"step": 493
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 25.974683544303797,
|
| 413 |
+
"grad_norm": 8.147224426269531,
|
| 414 |
+
"learning_rate": 4.055555555555556e-05,
|
| 415 |
+
"loss": 0.2806,
|
| 416 |
+
"step": 513
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 25.974683544303797,
|
| 420 |
+
"eval_accuracy": 0.8650793650793651,
|
| 421 |
+
"eval_loss": 0.4862901568412781,
|
| 422 |
+
"eval_runtime": 10.7246,
|
| 423 |
+
"eval_samples_per_second": 117.487,
|
| 424 |
+
"eval_steps_per_second": 0.932,
|
| 425 |
+
"step": 513
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 26.9873417721519,
|
| 429 |
+
"grad_norm": 1.3954640626907349,
|
| 430 |
+
"learning_rate": 3.9970760233918134e-05,
|
| 431 |
+
"loss": 0.2696,
|
| 432 |
+
"step": 533
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 26.9873417721519,
|
| 436 |
+
"eval_accuracy": 0.8595238095238096,
|
| 437 |
+
"eval_loss": 0.5064178705215454,
|
| 438 |
+
"eval_runtime": 10.7667,
|
| 439 |
+
"eval_samples_per_second": 117.027,
|
| 440 |
+
"eval_steps_per_second": 0.929,
|
| 441 |
+
"step": 533
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 28.0,
|
| 445 |
+
"grad_norm": 1.5532612800598145,
|
| 446 |
+
"learning_rate": 3.93859649122807e-05,
|
| 447 |
+
"loss": 0.2709,
|
| 448 |
+
"step": 553
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 28.0,
|
| 452 |
+
"eval_accuracy": 0.8674603174603175,
|
| 453 |
+
"eval_loss": 0.4656953811645508,
|
| 454 |
+
"eval_runtime": 10.8334,
|
| 455 |
+
"eval_samples_per_second": 116.307,
|
| 456 |
+
"eval_steps_per_second": 0.923,
|
| 457 |
+
"step": 553
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 28.962025316455698,
|
| 461 |
+
"grad_norm": 1.5748584270477295,
|
| 462 |
+
"learning_rate": 3.883040935672515e-05,
|
| 463 |
+
"loss": 0.2702,
|
| 464 |
+
"step": 572
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 28.962025316455698,
|
| 468 |
+
"eval_accuracy": 0.8571428571428571,
|
| 469 |
+
"eval_loss": 0.4933919310569763,
|
| 470 |
+
"eval_runtime": 10.7751,
|
| 471 |
+
"eval_samples_per_second": 116.936,
|
| 472 |
+
"eval_steps_per_second": 0.928,
|
| 473 |
+
"step": 572
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 29.974683544303797,
|
| 477 |
+
"grad_norm": 1.3018436431884766,
|
| 478 |
+
"learning_rate": 3.824561403508773e-05,
|
| 479 |
+
"loss": 0.2628,
|
| 480 |
+
"step": 592
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 29.974683544303797,
|
| 484 |
+
"eval_accuracy": 0.8555555555555555,
|
| 485 |
+
"eval_loss": 0.4940374493598938,
|
| 486 |
+
"eval_runtime": 10.7573,
|
| 487 |
+
"eval_samples_per_second": 117.13,
|
| 488 |
+
"eval_steps_per_second": 0.93,
|
| 489 |
+
"step": 592
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 30.9873417721519,
|
| 493 |
+
"grad_norm": 1.811011791229248,
|
| 494 |
+
"learning_rate": 3.7660818713450294e-05,
|
| 495 |
+
"loss": 0.2543,
|
| 496 |
+
"step": 612
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 30.9873417721519,
|
| 500 |
+
"eval_accuracy": 0.8642857142857143,
|
| 501 |
+
"eval_loss": 0.48308396339416504,
|
| 502 |
+
"eval_runtime": 10.8262,
|
| 503 |
+
"eval_samples_per_second": 116.384,
|
| 504 |
+
"eval_steps_per_second": 0.924,
|
| 505 |
+
"step": 612
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 32.0,
|
| 509 |
+
"grad_norm": 1.4332562685012817,
|
| 510 |
+
"learning_rate": 3.707602339181287e-05,
|
| 511 |
+
"loss": 0.2427,
|
| 512 |
+
"step": 632
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 32.0,
|
| 516 |
+
"eval_accuracy": 0.861904761904762,
|
| 517 |
+
"eval_loss": 0.4981466233730316,
|
| 518 |
+
"eval_runtime": 10.8291,
|
| 519 |
+
"eval_samples_per_second": 116.353,
|
| 520 |
+
"eval_steps_per_second": 0.923,
|
| 521 |
+
"step": 632
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 32.962025316455694,
|
| 525 |
+
"grad_norm": 1.5480865240097046,
|
| 526 |
+
"learning_rate": 3.652046783625731e-05,
|
| 527 |
+
"loss": 0.2659,
|
| 528 |
+
"step": 651
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 32.962025316455694,
|
| 532 |
+
"eval_accuracy": 0.8642857142857143,
|
| 533 |
+
"eval_loss": 0.5094291567802429,
|
| 534 |
+
"eval_runtime": 10.7442,
|
| 535 |
+
"eval_samples_per_second": 117.272,
|
| 536 |
+
"eval_steps_per_second": 0.931,
|
| 537 |
+
"step": 651
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 33.9746835443038,
|
| 541 |
+
"grad_norm": 1.3587052822113037,
|
| 542 |
+
"learning_rate": 3.593567251461988e-05,
|
| 543 |
+
"loss": 0.2398,
|
| 544 |
+
"step": 671
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 33.9746835443038,
|
| 548 |
+
"eval_accuracy": 0.8658730158730159,
|
| 549 |
+
"eval_loss": 0.5013704895973206,
|
| 550 |
+
"eval_runtime": 10.759,
|
| 551 |
+
"eval_samples_per_second": 117.112,
|
| 552 |
+
"eval_steps_per_second": 0.929,
|
| 553 |
+
"step": 671
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 34.9873417721519,
|
| 557 |
+
"grad_norm": 1.3286776542663574,
|
| 558 |
+
"learning_rate": 3.5350877192982455e-05,
|
| 559 |
+
"loss": 0.227,
|
| 560 |
+
"step": 691
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 34.9873417721519,
|
| 564 |
+
"eval_accuracy": 0.8634920634920635,
|
| 565 |
+
"eval_loss": 0.5037477612495422,
|
| 566 |
+
"eval_runtime": 10.7589,
|
| 567 |
+
"eval_samples_per_second": 117.112,
|
| 568 |
+
"eval_steps_per_second": 0.929,
|
| 569 |
+
"step": 691
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 36.0,
|
| 573 |
+
"grad_norm": 1.6547776460647583,
|
| 574 |
+
"learning_rate": 3.476608187134503e-05,
|
| 575 |
+
"loss": 0.2308,
|
| 576 |
+
"step": 711
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 36.0,
|
| 580 |
+
"eval_accuracy": 0.8658730158730159,
|
| 581 |
+
"eval_loss": 0.5068167448043823,
|
| 582 |
+
"eval_runtime": 10.7754,
|
| 583 |
+
"eval_samples_per_second": 116.933,
|
| 584 |
+
"eval_steps_per_second": 0.928,
|
| 585 |
+
"step": 711
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 36.962025316455694,
|
| 589 |
+
"grad_norm": 1.3685321807861328,
|
| 590 |
+
"learning_rate": 3.421052631578947e-05,
|
| 591 |
+
"loss": 0.2326,
|
| 592 |
+
"step": 730
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 36.962025316455694,
|
| 596 |
+
"eval_accuracy": 0.8650793650793651,
|
| 597 |
+
"eval_loss": 0.4980192184448242,
|
| 598 |
+
"eval_runtime": 10.761,
|
| 599 |
+
"eval_samples_per_second": 117.09,
|
| 600 |
+
"eval_steps_per_second": 0.929,
|
| 601 |
+
"step": 730
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 37.9746835443038,
|
| 605 |
+
"grad_norm": 1.2418972253799438,
|
| 606 |
+
"learning_rate": 3.362573099415205e-05,
|
| 607 |
+
"loss": 0.2242,
|
| 608 |
+
"step": 750
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 37.9746835443038,
|
| 612 |
+
"eval_accuracy": 0.8587301587301587,
|
| 613 |
+
"eval_loss": 0.4938106834888458,
|
| 614 |
+
"eval_runtime": 11.0548,
|
| 615 |
+
"eval_samples_per_second": 113.978,
|
| 616 |
+
"eval_steps_per_second": 0.905,
|
| 617 |
+
"step": 750
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 38.9873417721519,
|
| 621 |
+
"grad_norm": 1.3450112342834473,
|
| 622 |
+
"learning_rate": 3.304093567251462e-05,
|
| 623 |
+
"loss": 0.2152,
|
| 624 |
+
"step": 770
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 38.9873417721519,
|
| 628 |
+
"eval_accuracy": 0.8626984126984127,
|
| 629 |
+
"eval_loss": 0.49911221861839294,
|
| 630 |
+
"eval_runtime": 10.8459,
|
| 631 |
+
"eval_samples_per_second": 116.173,
|
| 632 |
+
"eval_steps_per_second": 0.922,
|
| 633 |
+
"step": 770
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 40.0,
|
| 637 |
+
"grad_norm": 1.3505226373672485,
|
| 638 |
+
"learning_rate": 3.24561403508772e-05,
|
| 639 |
+
"loss": 0.2205,
|
| 640 |
+
"step": 790
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 40.0,
|
| 644 |
+
"eval_accuracy": 0.8571428571428571,
|
| 645 |
+
"eval_loss": 0.5294431447982788,
|
| 646 |
+
"eval_runtime": 11.024,
|
| 647 |
+
"eval_samples_per_second": 114.296,
|
| 648 |
+
"eval_steps_per_second": 0.907,
|
| 649 |
+
"step": 790
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 40.962025316455694,
|
| 653 |
+
"grad_norm": 1.1549227237701416,
|
| 654 |
+
"learning_rate": 3.1900584795321634e-05,
|
| 655 |
+
"loss": 0.2299,
|
| 656 |
+
"step": 809
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 40.962025316455694,
|
| 660 |
+
"eval_accuracy": 0.8650793650793651,
|
| 661 |
+
"eval_loss": 0.5079935789108276,
|
| 662 |
+
"eval_runtime": 10.8051,
|
| 663 |
+
"eval_samples_per_second": 116.612,
|
| 664 |
+
"eval_steps_per_second": 0.925,
|
| 665 |
+
"step": 809
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 41.9746835443038,
|
| 669 |
+
"grad_norm": 1.252756118774414,
|
| 670 |
+
"learning_rate": 3.131578947368421e-05,
|
| 671 |
+
"loss": 0.1978,
|
| 672 |
+
"step": 829
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 41.9746835443038,
|
| 676 |
+
"eval_accuracy": 0.861904761904762,
|
| 677 |
+
"eval_loss": 0.5043396949768066,
|
| 678 |
+
"eval_runtime": 11.0945,
|
| 679 |
+
"eval_samples_per_second": 113.57,
|
| 680 |
+
"eval_steps_per_second": 0.901,
|
| 681 |
+
"step": 829
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 42.9873417721519,
|
| 685 |
+
"grad_norm": 1.20892333984375,
|
| 686 |
+
"learning_rate": 3.073099415204678e-05,
|
| 687 |
+
"loss": 0.2081,
|
| 688 |
+
"step": 849
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 42.9873417721519,
|
| 692 |
+
"eval_accuracy": 0.8634920634920635,
|
| 693 |
+
"eval_loss": 0.5008840560913086,
|
| 694 |
+
"eval_runtime": 10.8249,
|
| 695 |
+
"eval_samples_per_second": 116.398,
|
| 696 |
+
"eval_steps_per_second": 0.924,
|
| 697 |
+
"step": 849
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 44.0,
|
| 701 |
+
"grad_norm": 0.9471483826637268,
|
| 702 |
+
"learning_rate": 3.0146198830409357e-05,
|
| 703 |
+
"loss": 0.1893,
|
| 704 |
+
"step": 869
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 44.0,
|
| 708 |
+
"eval_accuracy": 0.8571428571428571,
|
| 709 |
+
"eval_loss": 0.5212369561195374,
|
| 710 |
+
"eval_runtime": 10.7573,
|
| 711 |
+
"eval_samples_per_second": 117.13,
|
| 712 |
+
"eval_steps_per_second": 0.93,
|
| 713 |
+
"step": 869
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 44.962025316455694,
|
| 717 |
+
"grad_norm": 1.4700783491134644,
|
| 718 |
+
"learning_rate": 2.95906432748538e-05,
|
| 719 |
+
"loss": 0.1988,
|
| 720 |
+
"step": 888
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 44.962025316455694,
|
| 724 |
+
"eval_accuracy": 0.8626984126984127,
|
| 725 |
+
"eval_loss": 0.4991550147533417,
|
| 726 |
+
"eval_runtime": 10.8133,
|
| 727 |
+
"eval_samples_per_second": 116.523,
|
| 728 |
+
"eval_steps_per_second": 0.925,
|
| 729 |
+
"step": 888
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 45.9746835443038,
|
| 733 |
+
"grad_norm": 1.0916502475738525,
|
| 734 |
+
"learning_rate": 2.9005847953216375e-05,
|
| 735 |
+
"loss": 0.1911,
|
| 736 |
+
"step": 908
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 45.9746835443038,
|
| 740 |
+
"eval_accuracy": 0.8674603174603175,
|
| 741 |
+
"eval_loss": 0.5237799882888794,
|
| 742 |
+
"eval_runtime": 10.7538,
|
| 743 |
+
"eval_samples_per_second": 117.168,
|
| 744 |
+
"eval_steps_per_second": 0.93,
|
| 745 |
+
"step": 908
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 46.9873417721519,
|
| 749 |
+
"grad_norm": 1.2590000629425049,
|
| 750 |
+
"learning_rate": 2.842105263157895e-05,
|
| 751 |
+
"loss": 0.1877,
|
| 752 |
+
"step": 928
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 46.9873417721519,
|
| 756 |
+
"eval_accuracy": 0.8674603174603175,
|
| 757 |
+
"eval_loss": 0.5184463262557983,
|
| 758 |
+
"eval_runtime": 10.8005,
|
| 759 |
+
"eval_samples_per_second": 116.662,
|
| 760 |
+
"eval_steps_per_second": 0.926,
|
| 761 |
+
"step": 928
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 48.0,
|
| 765 |
+
"grad_norm": 1.417845606803894,
|
| 766 |
+
"learning_rate": 2.783625730994152e-05,
|
| 767 |
+
"loss": 0.1957,
|
| 768 |
+
"step": 948
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 48.0,
|
| 772 |
+
"eval_accuracy": 0.8571428571428571,
|
| 773 |
+
"eval_loss": 0.5155279040336609,
|
| 774 |
+
"eval_runtime": 10.7518,
|
| 775 |
+
"eval_samples_per_second": 117.19,
|
| 776 |
+
"eval_steps_per_second": 0.93,
|
| 777 |
+
"step": 948
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 48.962025316455694,
|
| 781 |
+
"grad_norm": 1.304579496383667,
|
| 782 |
+
"learning_rate": 2.7280701754385968e-05,
|
| 783 |
+
"loss": 0.185,
|
| 784 |
+
"step": 967
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 48.962025316455694,
|
| 788 |
+
"eval_accuracy": 0.8674603174603175,
|
| 789 |
+
"eval_loss": 0.5028470158576965,
|
| 790 |
+
"eval_runtime": 10.7385,
|
| 791 |
+
"eval_samples_per_second": 117.335,
|
| 792 |
+
"eval_steps_per_second": 0.931,
|
| 793 |
+
"step": 967
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 49.9746835443038,
|
| 797 |
+
"grad_norm": 1.1883121728897095,
|
| 798 |
+
"learning_rate": 2.669590643274854e-05,
|
| 799 |
+
"loss": 0.1821,
|
| 800 |
+
"step": 987
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 49.9746835443038,
|
| 804 |
+
"eval_accuracy": 0.8626984126984127,
|
| 805 |
+
"eval_loss": 0.5118417739868164,
|
| 806 |
+
"eval_runtime": 10.7974,
|
| 807 |
+
"eval_samples_per_second": 116.694,
|
| 808 |
+
"eval_steps_per_second": 0.926,
|
| 809 |
+
"step": 987
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 50.9873417721519,
|
| 813 |
+
"grad_norm": 0.9844208359718323,
|
| 814 |
+
"learning_rate": 2.6111111111111114e-05,
|
| 815 |
+
"loss": 0.1843,
|
| 816 |
+
"step": 1007
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 50.9873417721519,
|
| 820 |
+
"eval_accuracy": 0.8650793650793651,
|
| 821 |
+
"eval_loss": 0.5006617903709412,
|
| 822 |
+
"eval_runtime": 10.7816,
|
| 823 |
+
"eval_samples_per_second": 116.866,
|
| 824 |
+
"eval_steps_per_second": 0.928,
|
| 825 |
+
"step": 1007
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 52.0,
|
| 829 |
+
"grad_norm": 1.392893671989441,
|
| 830 |
+
"learning_rate": 2.5526315789473688e-05,
|
| 831 |
+
"loss": 0.1711,
|
| 832 |
+
"step": 1027
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 52.0,
|
| 836 |
+
"eval_accuracy": 0.8571428571428571,
|
| 837 |
+
"eval_loss": 0.511702299118042,
|
| 838 |
+
"eval_runtime": 10.8792,
|
| 839 |
+
"eval_samples_per_second": 115.817,
|
| 840 |
+
"eval_steps_per_second": 0.919,
|
| 841 |
+
"step": 1027
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 52.962025316455694,
|
| 845 |
+
"grad_norm": 1.6035434007644653,
|
| 846 |
+
"learning_rate": 2.4970760233918132e-05,
|
| 847 |
+
"loss": 0.1903,
|
| 848 |
+
"step": 1046
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 52.962025316455694,
|
| 852 |
+
"eval_accuracy": 0.8587301587301587,
|
| 853 |
+
"eval_loss": 0.507360577583313,
|
| 854 |
+
"eval_runtime": 10.9028,
|
| 855 |
+
"eval_samples_per_second": 115.567,
|
| 856 |
+
"eval_steps_per_second": 0.917,
|
| 857 |
+
"step": 1046
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 53.9746835443038,
|
| 861 |
+
"grad_norm": 1.0503844022750854,
|
| 862 |
+
"learning_rate": 2.4385964912280703e-05,
|
| 863 |
+
"loss": 0.1713,
|
| 864 |
+
"step": 1066
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 53.9746835443038,
|
| 868 |
+
"eval_accuracy": 0.8658730158730159,
|
| 869 |
+
"eval_loss": 0.5167327523231506,
|
| 870 |
+
"eval_runtime": 10.8055,
|
| 871 |
+
"eval_samples_per_second": 116.608,
|
| 872 |
+
"eval_steps_per_second": 0.925,
|
| 873 |
+
"step": 1066
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 54.9873417721519,
|
| 877 |
+
"grad_norm": 1.0421777963638306,
|
| 878 |
+
"learning_rate": 2.3801169590643278e-05,
|
| 879 |
+
"loss": 0.1677,
|
| 880 |
+
"step": 1086
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 54.9873417721519,
|
| 884 |
+
"eval_accuracy": 0.8666666666666667,
|
| 885 |
+
"eval_loss": 0.5178954601287842,
|
| 886 |
+
"eval_runtime": 10.802,
|
| 887 |
+
"eval_samples_per_second": 116.645,
|
| 888 |
+
"eval_steps_per_second": 0.926,
|
| 889 |
+
"step": 1086
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 56.0,
|
| 893 |
+
"grad_norm": 1.283031940460205,
|
| 894 |
+
"learning_rate": 2.321637426900585e-05,
|
| 895 |
+
"loss": 0.16,
|
| 896 |
+
"step": 1106
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 56.0,
|
| 900 |
+
"eval_accuracy": 0.8650793650793651,
|
| 901 |
+
"eval_loss": 0.5145161747932434,
|
| 902 |
+
"eval_runtime": 10.7346,
|
| 903 |
+
"eval_samples_per_second": 117.377,
|
| 904 |
+
"eval_steps_per_second": 0.932,
|
| 905 |
+
"step": 1106
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 56.962025316455694,
|
| 909 |
+
"grad_norm": 2.3518636226654053,
|
| 910 |
+
"learning_rate": 2.2660818713450292e-05,
|
| 911 |
+
"loss": 0.1818,
|
| 912 |
+
"step": 1125
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 56.962025316455694,
|
| 916 |
+
"eval_accuracy": 0.8650793650793651,
|
| 917 |
+
"eval_loss": 0.5295758247375488,
|
| 918 |
+
"eval_runtime": 10.7992,
|
| 919 |
+
"eval_samples_per_second": 116.676,
|
| 920 |
+
"eval_steps_per_second": 0.926,
|
| 921 |
+
"step": 1125
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 57.9746835443038,
|
| 925 |
+
"grad_norm": 1.4039283990859985,
|
| 926 |
+
"learning_rate": 2.2076023391812867e-05,
|
| 927 |
+
"loss": 0.1725,
|
| 928 |
+
"step": 1145
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 57.9746835443038,
|
| 932 |
+
"eval_accuracy": 0.8642857142857143,
|
| 933 |
+
"eval_loss": 0.531140923500061,
|
| 934 |
+
"eval_runtime": 10.7624,
|
| 935 |
+
"eval_samples_per_second": 117.075,
|
| 936 |
+
"eval_steps_per_second": 0.929,
|
| 937 |
+
"step": 1145
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 58.9873417721519,
|
| 941 |
+
"grad_norm": 1.1663118600845337,
|
| 942 |
+
"learning_rate": 2.149122807017544e-05,
|
| 943 |
+
"loss": 0.1642,
|
| 944 |
+
"step": 1165
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 58.9873417721519,
|
| 948 |
+
"eval_accuracy": 0.8626984126984127,
|
| 949 |
+
"eval_loss": 0.5316585302352905,
|
| 950 |
+
"eval_runtime": 10.7932,
|
| 951 |
+
"eval_samples_per_second": 116.74,
|
| 952 |
+
"eval_steps_per_second": 0.927,
|
| 953 |
+
"step": 1165
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 60.0,
|
| 957 |
+
"grad_norm": 1.2489556074142456,
|
| 958 |
+
"learning_rate": 2.0906432748538013e-05,
|
| 959 |
+
"loss": 0.1626,
|
| 960 |
+
"step": 1185
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 60.0,
|
| 964 |
+
"eval_accuracy": 0.861904761904762,
|
| 965 |
+
"eval_loss": 0.5282865166664124,
|
| 966 |
+
"eval_runtime": 10.7244,
|
| 967 |
+
"eval_samples_per_second": 117.49,
|
| 968 |
+
"eval_steps_per_second": 0.932,
|
| 969 |
+
"step": 1185
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 60.962025316455694,
|
| 973 |
+
"grad_norm": 1.4613455533981323,
|
| 974 |
+
"learning_rate": 2.0350877192982456e-05,
|
| 975 |
+
"loss": 0.1621,
|
| 976 |
+
"step": 1204
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 60.962025316455694,
|
| 980 |
+
"eval_accuracy": 0.8603174603174604,
|
| 981 |
+
"eval_loss": 0.5266717076301575,
|
| 982 |
+
"eval_runtime": 10.8467,
|
| 983 |
+
"eval_samples_per_second": 116.164,
|
| 984 |
+
"eval_steps_per_second": 0.922,
|
| 985 |
+
"step": 1204
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 61.9746835443038,
|
| 989 |
+
"grad_norm": 1.4865529537200928,
|
| 990 |
+
"learning_rate": 1.976608187134503e-05,
|
| 991 |
+
"loss": 0.1503,
|
| 992 |
+
"step": 1224
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 61.9746835443038,
|
| 996 |
+
"eval_accuracy": 0.861904761904762,
|
| 997 |
+
"eval_loss": 0.5389307141304016,
|
| 998 |
+
"eval_runtime": 10.7859,
|
| 999 |
+
"eval_samples_per_second": 116.819,
|
| 1000 |
+
"eval_steps_per_second": 0.927,
|
| 1001 |
+
"step": 1224
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 62.9873417721519,
|
| 1005 |
+
"grad_norm": 1.0126421451568604,
|
| 1006 |
+
"learning_rate": 1.9181286549707602e-05,
|
| 1007 |
+
"loss": 0.162,
|
| 1008 |
+
"step": 1244
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 62.9873417721519,
|
| 1012 |
+
"eval_accuracy": 0.8626984126984127,
|
| 1013 |
+
"eval_loss": 0.540341317653656,
|
| 1014 |
+
"eval_runtime": 10.7503,
|
| 1015 |
+
"eval_samples_per_second": 117.206,
|
| 1016 |
+
"eval_steps_per_second": 0.93,
|
| 1017 |
+
"step": 1244
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 64.0,
|
| 1021 |
+
"grad_norm": 1.3681743144989014,
|
| 1022 |
+
"learning_rate": 1.8596491228070176e-05,
|
| 1023 |
+
"loss": 0.154,
|
| 1024 |
+
"step": 1264
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 64.0,
|
| 1028 |
+
"eval_accuracy": 0.8650793650793651,
|
| 1029 |
+
"eval_loss": 0.5240360498428345,
|
| 1030 |
+
"eval_runtime": 10.6902,
|
| 1031 |
+
"eval_samples_per_second": 117.865,
|
| 1032 |
+
"eval_steps_per_second": 0.935,
|
| 1033 |
+
"step": 1264
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 64.9620253164557,
|
| 1037 |
+
"grad_norm": 1.1314650774002075,
|
| 1038 |
+
"learning_rate": 1.804093567251462e-05,
|
| 1039 |
+
"loss": 0.1525,
|
| 1040 |
+
"step": 1283
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 64.9620253164557,
|
| 1044 |
+
"eval_accuracy": 0.8650793650793651,
|
| 1045 |
+
"eval_loss": 0.5337327718734741,
|
| 1046 |
+
"eval_runtime": 10.5876,
|
| 1047 |
+
"eval_samples_per_second": 119.007,
|
| 1048 |
+
"eval_steps_per_second": 0.945,
|
| 1049 |
+
"step": 1283
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 65.9746835443038,
|
| 1053 |
+
"grad_norm": 1.1210103034973145,
|
| 1054 |
+
"learning_rate": 1.745614035087719e-05,
|
| 1055 |
+
"loss": 0.1529,
|
| 1056 |
+
"step": 1303
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 65.9746835443038,
|
| 1060 |
+
"eval_accuracy": 0.8642857142857143,
|
| 1061 |
+
"eval_loss": 0.5457757115364075,
|
| 1062 |
+
"eval_runtime": 10.8103,
|
| 1063 |
+
"eval_samples_per_second": 116.555,
|
| 1064 |
+
"eval_steps_per_second": 0.925,
|
| 1065 |
+
"step": 1303
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 66.9873417721519,
|
| 1069 |
+
"grad_norm": 1.2637122869491577,
|
| 1070 |
+
"learning_rate": 1.6871345029239766e-05,
|
| 1071 |
+
"loss": 0.1548,
|
| 1072 |
+
"step": 1323
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 66.9873417721519,
|
| 1076 |
+
"eval_accuracy": 0.8650793650793651,
|
| 1077 |
+
"eval_loss": 0.5383771061897278,
|
| 1078 |
+
"eval_runtime": 10.877,
|
| 1079 |
+
"eval_samples_per_second": 115.84,
|
| 1080 |
+
"eval_steps_per_second": 0.919,
|
| 1081 |
+
"step": 1323
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 68.0,
|
| 1085 |
+
"grad_norm": 1.1812046766281128,
|
| 1086 |
+
"learning_rate": 1.628654970760234e-05,
|
| 1087 |
+
"loss": 0.1556,
|
| 1088 |
+
"step": 1343
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 68.0,
|
| 1092 |
+
"eval_accuracy": 0.8626984126984127,
|
| 1093 |
+
"eval_loss": 0.5395429134368896,
|
| 1094 |
+
"eval_runtime": 10.829,
|
| 1095 |
+
"eval_samples_per_second": 116.354,
|
| 1096 |
+
"eval_steps_per_second": 0.923,
|
| 1097 |
+
"step": 1343
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 68.9620253164557,
|
| 1101 |
+
"grad_norm": 1.21077299118042,
|
| 1102 |
+
"learning_rate": 1.5730994152046784e-05,
|
| 1103 |
+
"loss": 0.1629,
|
| 1104 |
+
"step": 1362
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 68.9620253164557,
|
| 1108 |
+
"eval_accuracy": 0.8634920634920635,
|
| 1109 |
+
"eval_loss": 0.5454122424125671,
|
| 1110 |
+
"eval_runtime": 10.6544,
|
| 1111 |
+
"eval_samples_per_second": 118.261,
|
| 1112 |
+
"eval_steps_per_second": 0.939,
|
| 1113 |
+
"step": 1362
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 69.9746835443038,
|
| 1117 |
+
"grad_norm": 1.3317054510116577,
|
| 1118 |
+
"learning_rate": 1.5146198830409358e-05,
|
| 1119 |
+
"loss": 0.1387,
|
| 1120 |
+
"step": 1382
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 69.9746835443038,
|
| 1124 |
+
"eval_accuracy": 0.8746031746031746,
|
| 1125 |
+
"eval_loss": 0.5267017483711243,
|
| 1126 |
+
"eval_runtime": 10.8484,
|
| 1127 |
+
"eval_samples_per_second": 116.147,
|
| 1128 |
+
"eval_steps_per_second": 0.922,
|
| 1129 |
+
"step": 1382
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 70.9873417721519,
|
| 1133 |
+
"grad_norm": 1.2003090381622314,
|
| 1134 |
+
"learning_rate": 1.4561403508771931e-05,
|
| 1135 |
+
"loss": 0.1495,
|
| 1136 |
+
"step": 1402
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 70.9873417721519,
|
| 1140 |
+
"eval_accuracy": 0.8650793650793651,
|
| 1141 |
+
"eval_loss": 0.5427414774894714,
|
| 1142 |
+
"eval_runtime": 10.7136,
|
| 1143 |
+
"eval_samples_per_second": 117.608,
|
| 1144 |
+
"eval_steps_per_second": 0.933,
|
| 1145 |
+
"step": 1402
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 72.0,
|
| 1149 |
+
"grad_norm": 1.2609037160873413,
|
| 1150 |
+
"learning_rate": 1.3976608187134504e-05,
|
| 1151 |
+
"loss": 0.1465,
|
| 1152 |
+
"step": 1422
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 72.0,
|
| 1156 |
+
"eval_accuracy": 0.8690476190476191,
|
| 1157 |
+
"eval_loss": 0.559054434299469,
|
| 1158 |
+
"eval_runtime": 10.7661,
|
| 1159 |
+
"eval_samples_per_second": 117.034,
|
| 1160 |
+
"eval_steps_per_second": 0.929,
|
| 1161 |
+
"step": 1422
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 72.9620253164557,
|
| 1165 |
+
"grad_norm": 1.456437587738037,
|
| 1166 |
+
"learning_rate": 1.3421052631578948e-05,
|
| 1167 |
+
"loss": 0.1478,
|
| 1168 |
+
"step": 1441
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 72.9620253164557,
|
| 1172 |
+
"eval_accuracy": 0.8722222222222222,
|
| 1173 |
+
"eval_loss": 0.532349705696106,
|
| 1174 |
+
"eval_runtime": 10.7627,
|
| 1175 |
+
"eval_samples_per_second": 117.071,
|
| 1176 |
+
"eval_steps_per_second": 0.929,
|
| 1177 |
+
"step": 1441
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 73.9746835443038,
|
| 1181 |
+
"grad_norm": 1.404703140258789,
|
| 1182 |
+
"learning_rate": 1.283625730994152e-05,
|
| 1183 |
+
"loss": 0.1447,
|
| 1184 |
+
"step": 1461
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 73.9746835443038,
|
| 1188 |
+
"eval_accuracy": 0.8690476190476191,
|
| 1189 |
+
"eval_loss": 0.5560940504074097,
|
| 1190 |
+
"eval_runtime": 10.7443,
|
| 1191 |
+
"eval_samples_per_second": 117.271,
|
| 1192 |
+
"eval_steps_per_second": 0.931,
|
| 1193 |
+
"step": 1461
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 74.9873417721519,
|
| 1197 |
+
"grad_norm": 1.3342186212539673,
|
| 1198 |
+
"learning_rate": 1.2251461988304095e-05,
|
| 1199 |
+
"loss": 0.1435,
|
| 1200 |
+
"step": 1481
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 74.9873417721519,
|
| 1204 |
+
"eval_accuracy": 0.8658730158730159,
|
| 1205 |
+
"eval_loss": 0.5346001982688904,
|
| 1206 |
+
"eval_runtime": 10.8529,
|
| 1207 |
+
"eval_samples_per_second": 116.098,
|
| 1208 |
+
"eval_steps_per_second": 0.921,
|
| 1209 |
+
"step": 1481
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 76.0,
|
| 1213 |
+
"grad_norm": 1.2481029033660889,
|
| 1214 |
+
"learning_rate": 1.1666666666666668e-05,
|
| 1215 |
+
"loss": 0.1459,
|
| 1216 |
+
"step": 1501
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 76.0,
|
| 1220 |
+
"eval_accuracy": 0.8658730158730159,
|
| 1221 |
+
"eval_loss": 0.5466868281364441,
|
| 1222 |
+
"eval_runtime": 11.0795,
|
| 1223 |
+
"eval_samples_per_second": 113.723,
|
| 1224 |
+
"eval_steps_per_second": 0.903,
|
| 1225 |
+
"step": 1501
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 76.9620253164557,
|
| 1229 |
+
"grad_norm": 1.2153362035751343,
|
| 1230 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 1231 |
+
"loss": 0.1474,
|
| 1232 |
+
"step": 1520
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 76.9620253164557,
|
| 1236 |
+
"eval_accuracy": 0.8690476190476191,
|
| 1237 |
+
"eval_loss": 0.5463184118270874,
|
| 1238 |
+
"eval_runtime": 10.7937,
|
| 1239 |
+
"eval_samples_per_second": 116.735,
|
| 1240 |
+
"eval_steps_per_second": 0.926,
|
| 1241 |
+
"step": 1520
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 77.9746835443038,
|
| 1245 |
+
"grad_norm": 1.2351834774017334,
|
| 1246 |
+
"learning_rate": 1.0526315789473684e-05,
|
| 1247 |
+
"loss": 0.1352,
|
| 1248 |
+
"step": 1540
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 77.9746835443038,
|
| 1252 |
+
"eval_accuracy": 0.8650793650793651,
|
| 1253 |
+
"eval_loss": 0.5412562489509583,
|
| 1254 |
+
"eval_runtime": 11.1033,
|
| 1255 |
+
"eval_samples_per_second": 113.48,
|
| 1256 |
+
"eval_steps_per_second": 0.901,
|
| 1257 |
+
"step": 1540
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 78.9873417721519,
|
| 1261 |
+
"grad_norm": 1.3961732387542725,
|
| 1262 |
+
"learning_rate": 9.941520467836257e-06,
|
| 1263 |
+
"loss": 0.1337,
|
| 1264 |
+
"step": 1560
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 78.9873417721519,
|
| 1268 |
+
"eval_accuracy": 0.8650793650793651,
|
| 1269 |
+
"eval_loss": 0.5488775372505188,
|
| 1270 |
+
"eval_runtime": 10.7671,
|
| 1271 |
+
"eval_samples_per_second": 117.023,
|
| 1272 |
+
"eval_steps_per_second": 0.929,
|
| 1273 |
+
"step": 1560
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 80.0,
|
| 1277 |
+
"grad_norm": 1.8050953149795532,
|
| 1278 |
+
"learning_rate": 9.35672514619883e-06,
|
| 1279 |
+
"loss": 0.1374,
|
| 1280 |
+
"step": 1580
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 80.0,
|
| 1284 |
+
"eval_accuracy": 0.8587301587301587,
|
| 1285 |
+
"eval_loss": 0.5454345941543579,
|
| 1286 |
+
"eval_runtime": 10.8299,
|
| 1287 |
+
"eval_samples_per_second": 116.345,
|
| 1288 |
+
"eval_steps_per_second": 0.923,
|
| 1289 |
+
"step": 1580
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 80.9620253164557,
|
| 1293 |
+
"grad_norm": 1.2362314462661743,
|
| 1294 |
+
"learning_rate": 8.801169590643275e-06,
|
| 1295 |
+
"loss": 0.1383,
|
| 1296 |
+
"step": 1599
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 80.9620253164557,
|
| 1300 |
+
"eval_accuracy": 0.8626984126984127,
|
| 1301 |
+
"eval_loss": 0.5432500243186951,
|
| 1302 |
+
"eval_runtime": 10.8767,
|
| 1303 |
+
"eval_samples_per_second": 115.844,
|
| 1304 |
+
"eval_steps_per_second": 0.919,
|
| 1305 |
+
"step": 1599
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 81.9746835443038,
|
| 1309 |
+
"grad_norm": 1.1372051239013672,
|
| 1310 |
+
"learning_rate": 8.216374269005848e-06,
|
| 1311 |
+
"loss": 0.1408,
|
| 1312 |
+
"step": 1619
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 81.9746835443038,
|
| 1316 |
+
"eval_accuracy": 0.8682539682539683,
|
| 1317 |
+
"eval_loss": 0.5383033156394958,
|
| 1318 |
+
"eval_runtime": 10.9264,
|
| 1319 |
+
"eval_samples_per_second": 115.317,
|
| 1320 |
+
"eval_steps_per_second": 0.915,
|
| 1321 |
+
"step": 1619
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 82.9873417721519,
|
| 1325 |
+
"grad_norm": 1.6927990913391113,
|
| 1326 |
+
"learning_rate": 7.631578947368421e-06,
|
| 1327 |
+
"loss": 0.134,
|
| 1328 |
+
"step": 1639
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 82.9873417721519,
|
| 1332 |
+
"eval_accuracy": 0.8642857142857143,
|
| 1333 |
+
"eval_loss": 0.5522441267967224,
|
| 1334 |
+
"eval_runtime": 10.7932,
|
| 1335 |
+
"eval_samples_per_second": 116.74,
|
| 1336 |
+
"eval_steps_per_second": 0.927,
|
| 1337 |
+
"step": 1639
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 84.0,
|
| 1341 |
+
"grad_norm": 1.1734745502471924,
|
| 1342 |
+
"learning_rate": 7.046783625730995e-06,
|
| 1343 |
+
"loss": 0.1353,
|
| 1344 |
+
"step": 1659
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 84.0,
|
| 1348 |
+
"eval_accuracy": 0.8579365079365079,
|
| 1349 |
+
"eval_loss": 0.5485585331916809,
|
| 1350 |
+
"eval_runtime": 10.8452,
|
| 1351 |
+
"eval_samples_per_second": 116.181,
|
| 1352 |
+
"eval_steps_per_second": 0.922,
|
| 1353 |
+
"step": 1659
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 84.9620253164557,
|
| 1357 |
+
"grad_norm": 1.3662621974945068,
|
| 1358 |
+
"learning_rate": 6.4912280701754385e-06,
|
| 1359 |
+
"loss": 0.1435,
|
| 1360 |
+
"step": 1678
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 84.9620253164557,
|
| 1364 |
+
"eval_accuracy": 0.8595238095238096,
|
| 1365 |
+
"eval_loss": 0.5582545399665833,
|
| 1366 |
+
"eval_runtime": 10.7527,
|
| 1367 |
+
"eval_samples_per_second": 117.18,
|
| 1368 |
+
"eval_steps_per_second": 0.93,
|
| 1369 |
+
"step": 1678
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 85.9746835443038,
|
| 1373 |
+
"grad_norm": 1.3297693729400635,
|
| 1374 |
+
"learning_rate": 5.906432748538012e-06,
|
| 1375 |
+
"loss": 0.1324,
|
| 1376 |
+
"step": 1698
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 85.9746835443038,
|
| 1380 |
+
"eval_accuracy": 0.861904761904762,
|
| 1381 |
+
"eval_loss": 0.5551320910453796,
|
| 1382 |
+
"eval_runtime": 10.8253,
|
| 1383 |
+
"eval_samples_per_second": 116.394,
|
| 1384 |
+
"eval_steps_per_second": 0.924,
|
| 1385 |
+
"step": 1698
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 86.9873417721519,
|
| 1389 |
+
"grad_norm": 1.2304210662841797,
|
| 1390 |
+
"learning_rate": 5.321637426900585e-06,
|
| 1391 |
+
"loss": 0.1306,
|
| 1392 |
+
"step": 1718
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 86.9873417721519,
|
| 1396 |
+
"eval_accuracy": 0.8611111111111112,
|
| 1397 |
+
"eval_loss": 0.553473949432373,
|
| 1398 |
+
"eval_runtime": 10.7756,
|
| 1399 |
+
"eval_samples_per_second": 116.931,
|
| 1400 |
+
"eval_steps_per_second": 0.928,
|
| 1401 |
+
"step": 1718
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 88.0,
|
| 1405 |
+
"grad_norm": 1.323527216911316,
|
| 1406 |
+
"learning_rate": 4.736842105263159e-06,
|
| 1407 |
+
"loss": 0.1348,
|
| 1408 |
+
"step": 1738
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 88.0,
|
| 1412 |
+
"eval_accuracy": 0.8666666666666667,
|
| 1413 |
+
"eval_loss": 0.5498299598693848,
|
| 1414 |
+
"eval_runtime": 10.7878,
|
| 1415 |
+
"eval_samples_per_second": 116.799,
|
| 1416 |
+
"eval_steps_per_second": 0.927,
|
| 1417 |
+
"step": 1738
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 88.9620253164557,
|
| 1421 |
+
"grad_norm": 1.0867611169815063,
|
| 1422 |
+
"learning_rate": 4.181286549707602e-06,
|
| 1423 |
+
"loss": 0.1334,
|
| 1424 |
+
"step": 1757
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 88.9620253164557,
|
| 1428 |
+
"eval_accuracy": 0.8658730158730159,
|
| 1429 |
+
"eval_loss": 0.5582374930381775,
|
| 1430 |
+
"eval_runtime": 10.7756,
|
| 1431 |
+
"eval_samples_per_second": 116.931,
|
| 1432 |
+
"eval_steps_per_second": 0.928,
|
| 1433 |
+
"step": 1757
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 89.9746835443038,
|
| 1437 |
+
"grad_norm": 1.0990999937057495,
|
| 1438 |
+
"learning_rate": 3.5964912280701756e-06,
|
| 1439 |
+
"loss": 0.1343,
|
| 1440 |
+
"step": 1777
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 89.9746835443038,
|
| 1444 |
+
"eval_accuracy": 0.8658730158730159,
|
| 1445 |
+
"eval_loss": 0.5526331067085266,
|
| 1446 |
+
"eval_runtime": 10.8124,
|
| 1447 |
+
"eval_samples_per_second": 116.533,
|
| 1448 |
+
"eval_steps_per_second": 0.925,
|
| 1449 |
+
"step": 1777
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 90.9873417721519,
|
| 1453 |
+
"grad_norm": 1.3471728563308716,
|
| 1454 |
+
"learning_rate": 3.011695906432749e-06,
|
| 1455 |
+
"loss": 0.1275,
|
| 1456 |
+
"step": 1797
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 90.9873417721519,
|
| 1460 |
+
"eval_accuracy": 0.8650793650793651,
|
| 1461 |
+
"eval_loss": 0.5543471574783325,
|
| 1462 |
+
"eval_runtime": 10.7534,
|
| 1463 |
+
"eval_samples_per_second": 117.172,
|
| 1464 |
+
"eval_steps_per_second": 0.93,
|
| 1465 |
+
"step": 1797
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 92.0,
|
| 1469 |
+
"grad_norm": 1.3125709295272827,
|
| 1470 |
+
"learning_rate": 2.426900584795322e-06,
|
| 1471 |
+
"loss": 0.1285,
|
| 1472 |
+
"step": 1817
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 92.0,
|
| 1476 |
+
"eval_accuracy": 0.8674603174603175,
|
| 1477 |
+
"eval_loss": 0.551249086856842,
|
| 1478 |
+
"eval_runtime": 10.7174,
|
| 1479 |
+
"eval_samples_per_second": 117.566,
|
| 1480 |
+
"eval_steps_per_second": 0.933,
|
| 1481 |
+
"step": 1817
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 92.9620253164557,
|
| 1485 |
+
"grad_norm": 1.069954752922058,
|
| 1486 |
+
"learning_rate": 1.8713450292397662e-06,
|
| 1487 |
+
"loss": 0.127,
|
| 1488 |
+
"step": 1836
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 92.9620253164557,
|
| 1492 |
+
"eval_accuracy": 0.8634920634920635,
|
| 1493 |
+
"eval_loss": 0.5507932305335999,
|
| 1494 |
+
"eval_runtime": 10.7352,
|
| 1495 |
+
"eval_samples_per_second": 117.371,
|
| 1496 |
+
"eval_steps_per_second": 0.932,
|
| 1497 |
+
"step": 1836
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 93.9746835443038,
|
| 1501 |
+
"grad_norm": 1.1107105016708374,
|
| 1502 |
+
"learning_rate": 1.2865497076023394e-06,
|
| 1503 |
+
"loss": 0.1258,
|
| 1504 |
+
"step": 1856
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 93.9746835443038,
|
| 1508 |
+
"eval_accuracy": 0.8642857142857143,
|
| 1509 |
+
"eval_loss": 0.5506840944290161,
|
| 1510 |
+
"eval_runtime": 10.5904,
|
| 1511 |
+
"eval_samples_per_second": 118.975,
|
| 1512 |
+
"eval_steps_per_second": 0.944,
|
| 1513 |
+
"step": 1856
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 94.9873417721519,
|
| 1517 |
+
"grad_norm": 1.1726576089859009,
|
| 1518 |
+
"learning_rate": 7.017543859649123e-07,
|
| 1519 |
+
"loss": 0.1119,
|
| 1520 |
+
"step": 1876
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 94.9873417721519,
|
| 1524 |
+
"eval_accuracy": 0.8666666666666667,
|
| 1525 |
+
"eval_loss": 0.5506576299667358,
|
| 1526 |
+
"eval_runtime": 10.8474,
|
| 1527 |
+
"eval_samples_per_second": 116.157,
|
| 1528 |
+
"eval_steps_per_second": 0.922,
|
| 1529 |
+
"step": 1876
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 96.0,
|
| 1533 |
+
"grad_norm": 1.4846915006637573,
|
| 1534 |
+
"learning_rate": 1.1695906432748539e-07,
|
| 1535 |
+
"loss": 0.1322,
|
| 1536 |
+
"step": 1896
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 96.0,
|
| 1540 |
+
"eval_accuracy": 0.8658730158730159,
|
| 1541 |
+
"eval_loss": 0.5504564046859741,
|
| 1542 |
+
"eval_runtime": 11.0992,
|
| 1543 |
+
"eval_samples_per_second": 113.522,
|
| 1544 |
+
"eval_steps_per_second": 0.901,
|
| 1545 |
+
"step": 1896
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 96.20253164556962,
|
| 1549 |
+
"grad_norm": 1.0216985940933228,
|
| 1550 |
+
"learning_rate": 0.0,
|
| 1551 |
+
"loss": 0.1315,
|
| 1552 |
+
"step": 1900
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 96.20253164556962,
|
| 1556 |
+
"eval_accuracy": 0.8658730158730159,
|
| 1557 |
+
"eval_loss": 0.5504307150840759,
|
| 1558 |
+
"eval_runtime": 10.802,
|
| 1559 |
+
"eval_samples_per_second": 116.645,
|
| 1560 |
+
"eval_steps_per_second": 0.926,
|
| 1561 |
+
"step": 1900
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 96.20253164556962,
|
| 1565 |
+
"step": 1900,
|
| 1566 |
+
"total_flos": 7.515490775048022e+19,
|
| 1567 |
+
"train_loss": 0.33647052476280614,
|
| 1568 |
+
"train_runtime": 20573.1873,
|
| 1569 |
+
"train_samples_per_second": 48.996,
|
| 1570 |
+
"train_steps_per_second": 0.092
|
| 1571 |
+
}
|
| 1572 |
+
],
|
| 1573 |
+
"logging_steps": 500,
|
| 1574 |
+
"max_steps": 1900,
|
| 1575 |
+
"num_input_tokens_seen": 0,
|
| 1576 |
+
"num_train_epochs": 100,
|
| 1577 |
+
"save_steps": 500,
|
| 1578 |
+
"stateful_callbacks": {
|
| 1579 |
+
"TrainerControl": {
|
| 1580 |
+
"args": {
|
| 1581 |
+
"should_epoch_stop": false,
|
| 1582 |
+
"should_evaluate": false,
|
| 1583 |
+
"should_log": false,
|
| 1584 |
+
"should_save": true,
|
| 1585 |
+
"should_training_stop": true
|
| 1586 |
+
},
|
| 1587 |
+
"attributes": {}
|
| 1588 |
+
}
|
| 1589 |
+
},
|
| 1590 |
+
"total_flos": 7.515490775048022e+19,
|
| 1591 |
+
"train_batch_size": 128,
|
| 1592 |
+
"trial_name": null,
|
| 1593 |
+
"trial_params": null
|
| 1594 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a1c0f33a3024e6ec45ce8978209c580f91e2084ba0bf40c70af9b63aea9815a
|
| 3 |
+
size 5112
|